Introduction

When Prof (Professor) Geesje van den Berg first informed me that they were working on a book about AI in education, I was thrilled, as discussions surrounding technological innovation are often dominated by the discourse of the Global North, meaning that the voices from the developing economies of the Global South are insufficiently represented. When she later invited me to contribute a foreword, I felt both honoured and intrigued to be considered, as there are many individuals undertaking pioneering work in this field who might offer a more unique perspective. However, Prof Van den Berg and I have known each other for many years, having collaborated at UNISA on an interim educator qualification called the National Professional Diploma in Education, and subsequently on a Master’s programme in open and distance learning (ODL), as well as having contributed to the same textbook on Curriculum Studies. Therefore, I was confident that the new publication would be something to eagerly anticipate.

I also have a longstanding interest in the use of technology, particularly, though not exclusively, in the provision of ODL. As noted in a recent keynote presentation for UNISA (Mays 2024a) several years ago, I worked with a large group of out-of-school youths aged around 23 who had completed their matriculation but were unable to pass a standard industry mathematics test required for entry-level employment opportunities. We needed a responsive online platform to motivate these learners to reengage with mathematics at approximately a Grade 6 level. We opted for Khan Academy, which was already utilising a form of AI (artificial intelligence). Students worked through activities and earned badges for motivation. A dashboard displayed their progress, showing how each small achievement contributed to completing the overall programme. Within minutes, each student was following a slightly different trajectory through the subject content in a manner that would have been difficult for a lone educator in a physical classroom using printed textbooks to replicate (there were more than 100 students in the group). Furthermore, the backend tutor functionality allowed me, as their chosen tutor, to monitor their progress and provide additional personalised feedback as needed. Most of the time, only a small percentage of students required this personalised human engagement. I was very impressed by how the technology enabled me to offer more tailored support.

Now that I am Director of the Education Sector at the COL (Commonwealth of Learning), that interest in technology continues. I believe that the COL has a key role to play in supporting the development and dissemination of useful knowledge in an open way to address practical, theoretical, and policy issues in the use of technology in education provision generally and with the evolving possibilities of Gen-AI (generative artificial intelligence) in particular. For example, as reported in our news stories, the COL recently collaborated with the National University of Samoa to use Gen-AI to develop a virtual tutor, able to answer most recurring questions about the Moodle platform being used (COL 2023). Subsequently, the COL worked with the University of the South Pacific to successfully enhance its Semester Zero induction programme, using similar technology to address both learning management systems and content queries (COL 2024). Learning from these experiences, COL recently inaugurated a collaborative project called ‘Teacher-in-the-Loop,’ in which several of our initiatives work together to use Gen-AI to help educators develop OER (open educational resources) for STEM (Science, Technology, Engineering, and Mathematics) and TVET (technical and vocational education and training) subjects, which is not only scientifically accurate but also actively seeks to pre-empt any possible biases, particularly in terms of gender equity. A key proposition of this ongoing project is that although Gen-AI is the tool, the educator makes the final decision on what to use and how.

The COL also recently published the results of a survey indicating that more support was needed for policy development in relation to AI deployment (Paskevicius 2024) – a need which was confirmed by other research (Linderoth, Hultén, & Stenliden 2024; Ratten & Jones 2023) – and followed up on the findings of this report by developing policy guidelines (Mohamed & Mishra 2024). So, our interests are aligned.

In the balance of this foreword, I want to reflect a little on the notion of human intelligence, the evolving field of AI, and some of the practical implications for teaching and learning, and then use these reflections as lenses for my initial engagement with this exciting new publication.

Reflection on Intelligence

The African proverb that heads this foreword offers a pragmatic insight into how to deal with change. This seems like a good starting point because, at least to my mind, change seems central to the concept of intelligence, whether reactive or proactive. Moreover, a publication focused on ‘artificial’ intelligence presupposes a contrast with ‘non-artificial,’ presumably human intelligence. Sternberg (2024), writing for the Encyclopedia Brittanica, offers the following definition: ‘[H]uman intelligence, mental quality that consists of the abilities to learn from experience, adapt to new situations, understand and handle abstract concepts, and use knowledge to manipulate one’s environment.’

Sternberg notes that there are various theories about the nature of human intelligence and identifies the following core groups, among others: Psychometric, cognitive, cognitive-contextual, biological, and hemispheric. He also observes that our intelligence/s may change over time. We do not have space to explore all these perspectives here, but it seems useful to include a short reflection to help explain the complexity. Human beings access information and experiences through their senses, so it is important for us to ask questions like, ‘Did I really see what I thought I saw?,’ ‘Did I really hear what I thought I heard?’ and so on. Once we have clarity on these questions, we tend to have an emotional response, for example, ‘I don’t like that, so I won’t engage further,’ while a more intelligent response might be, ‘I don’t like that; why do I not like it? What can I learn by trying to understand my dislike?’

In a similar vein, human beings live in a cultural context and might react to information or an experience by not wanting to engage further because what is said or done and by whom run counter to their own cultural norms. However, they might learn something by consciously examining their own norms and comparing them with those of others. Once we have passed through these initial filters, we might bring to bear our reasoning and logic to examine the information or experience in a critical way to understand it better, and then once we have a better understanding, we will likely evaluate that understanding from the perspective of our own worldview. The point being made is that human beings engage with the world in a multi-faceted way. They also make many daily decisions which are not necessarily based on logic or on an analysis of huge datasets and, in fact, human beings often infer meaning and make decisions based on very limited data. Moreover, human beings make connections between data, ideas, and phenomena that are sometimes not obviously connected at all and often also disparate in form. In addition, we have a strong propensity to imagine that which does not exist, or at least not yet in some cases; so authors like Jules Verne, Isaac Asimov, and Arther Clarke were able to write novels about journeys to the bottom of the sea or to the moon, and about robots and earth satellites, even before such things became possible technologically. Human intelligence (or intelligences) is therefore a complex field, but is the key counterpoint to the current focus on AI.

The evolving field of Artificial Intelligence

Oxford Languages (2024) defines AI as ‘the theory and development of computer systems able to perform tasks that normally require human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages.’ Although the concept of non-human intelligence can be found in ancient mythology and in a sense, began to seem practicable near the start of the industrial era with the invention of machines like Babbage’s difference engine, Scott (2024) observes that what we now refer to as AI has been around for a long time even in the modern era, with a focus on ‘programmed learning’ in the 1960s and 1970s, which advanced in the 1980s and 1990s into basic personalisation, the development of more adaptive virtual systems in the 2000s, and the more recent LLMs (large language models) revolution which has paved the way for virtual tutoring, chatbots for teletutoring, personalised learning content, automated grading, and automated administrative tasks. As with any new technology, there are potential opportunities, potential risks, and the need for experimentation and discourse (Baidoo-Anu & Owusu Ansah 2023). We also need to learn how best to use the evolving affordances of the technology, including gaining a nuanced understanding of its strengths and weaknesses in different contexts (Bozkurt & Sharma 2023; Cooper 2023; Du, Sun, Jiang, Atiquil Islam, & Gu 2024; Smolansky, Cram, Raduescu, Zeivots, Huber, & Kizilcec 2023).

As Denny, Gulwani, Heffernan, Käser, Moore, Rafferty, and Singla (2024) observe, the appropriate use of AI in an education context requires technical expertise, experience of working with students, and knowing how to deploy technology at scale. It will impact how we teach, design a curriculum, manage assessment, work with OER, and manage tensions such as between homogenisation, creativity, and the personalised/individualised. It will allow personalised, real-time feedback and create more time for human facilitation, but it also raises concerns about data privacy, algorithmic bias, academic integrity, and the need for substantially more educator training (Eke 2023; Nikolic, Daniel, Haque, Belkina, Hassan, & Grundy 2023; OER Africa 2023; Onesi-Ozigagun, Ololade, Eyo-Udo, & Ogundipe 2024).

In the most recent (7th) edition of their research on AI, Perrault and Clark (2024) conclude that AI beats humans on some tasks, but not all, that it can make workers more productive and lead to higher quality work, and also that it can accelerate scientific progress. However, they note the lack of robust and standardised evaluations for LLM responsibility and observe that as people become more aware of what AI can do, they also become more concerned about the implications. This would include concerns related to ethics, curriculum design, continuous learning, and industry alignment, among others (Abulibdeh, Zaidan, & Abulibdeh 2024).

Williamson (2024) observes that while we cannot anticipate all future risks, we know from our past experience that the deployment of new technologies is not a linear process. We therefore need an active discourse between the technology developers on the one hand and educators on the other to preempt any unintended consequences.

Overall, this suggests the need for a pragmatic response to use human intelligence and human-made intelligence in a synergistic way. However, some things already seem obvious to implement: We can automate routine administration and teaching tasks; manage a balance between personalised support and over-dependence; support students and educators to use and critique AI responses; focus on more authentic assessment; address cybersecurity and academic integrity issues; make ethical use of analytics; provide ongoing professional development for educators; and be open to trying new things in a responsible and ethical way (Bates 2024; Du et al. 2024; Kizilcec 2024; Law 2024; Sevnarayan & Potter 2024; Williamson 2024).

Implications for Practice

As noted in another discussion (Mays 2024b), we need to explore the link between pedagogy and technology and the implications for educators’ practice. Having established a framework, we can then explore the possible disruptive influence of AI. We also need to revisit assessment because if we change our assessment, then other things will fall into place (Rowntree 1987). In this context, we should think of pedagogy as an overarching term meaning the art and science of teaching, realising that it needs to be nuanced for different populations and contexts – for example, for adults (andragogy), for self-directed learning in ODL (heutagogy), and for appropriateness for cultural context (e.g., ubuntugogy from southern Africa). Overall, we should promote the notion of ‘open’ pedagogy (Bates 2022; Hegarty 2015).

We have no shortage of theories about how people learn, as the interactive map created by Millwood (2021) demonstrates. Despite this diversity of learning theories, we can identify at least four broad approaches: Behaviourist/instructivist, cognitive/ constructivist, socio-constructivist, and connectivist. These should not be regarded as mutually exclusive positions – our practice likely draws on all four to a lesser or greater extent at different times. Depending on our perspective, we will then use a particular technology in a different way, for example, the use of instructivist lecture-type videos in xMOOCs (extended massive open online courses) as opposed to the use of videos in cMOOCs (‘connectivist’ or ‘constructivist’ massive open online courses) where the video content will likely be scaffolded and presented differently but be augmented by additional applications like discussion forums. How we engage with AI or Gen-AI will similarly be influenced by what we believe about how learning happens.

Given the continuing challenges of the digital divide, we need to adopt a pragmatic approach to the use of technology. In many countries we still need to make provision for content that can be printed and/or shared digitally to a basic device at a Wi-Fi-enabled centre, especially for the more rural areas. We can, however, enrich the core content with, for example, video and audio content – whether using broadcast technology or embedded in an open textbook for students with higher end devices but limited access to the internet. We might then also need a version that is fully online, exploiting H5P (HTML5 package) for interactivity and including online discussion fora, which could have both synchronous and non-synchronous elements, as well as involving social media. As noted in the recently published GEM (global education monitoring) report for the Pacific (COL & UNESCO 2024), despite challenging contexts, we can and often do adapt technologies to find solutions that are more fit for context – AI and Gen-AI are no exception.

Once we know what we want to teach, why, and how, there is no shortage of technologies to choose from. The latest version of the ‘padagogy’ wheel, recently updated to include AI (Carrington 2024), provides some examples. However, we should remember the creator’s earlier observation that it is not fundamentally about the technology but the pedagogy (Carrington 2015). Bates (n.d.) offers a useful heuristic SECTIONS in this regard.

Stoker (2024), a LINC (logic and information network compiler) instructor for the ISS (international space station) of British Columbia, regularly uses Chat GPT (generative pre-trained transformer) in his work. He notes that it helps him to reduce the time needed for planning, assessment design, differentiation, and analysis of what is working in his presentations and what not, so that changes can be made before a topic is taught again and that over time, working with the application, it has learned what he likes and makes increasingly useful recommendations. Stoker also observes that AI has been used to support learning in multiple contexts ranging from Georgia State University, which in 2013 introduced AI to identify and automate feedback to at-risk students, linking them directly with a human tutor where indicated, to schools in rural India which developed AI tools for tablets to provide feedback to students on their subject-based learning. In both cases, these interventions resulted in improved retention and success. He notes that Singapore even experimented with humanoid robots to provide support for students whose educators were on sick leave. He notes that AI in the classroom can take care of many routine tasks freeing the educator to spend more time interacting directly with their students.

As with all new technologies, AI presents both opportunities and challenges. We should not, perhaps, succumb to the hype of the early adopters, but at the same, we should guard against being overly conservative and deny our educators and students the opportunity to exploit some of the more obvious early benefits. This publication can help us make more informed choices.

Overview of the Chapters in this Publication

This publication is divided into two volumes. In Volume 1, Chapter 1, Garth Aziz discusses the potential use of AI in pedagogies employed in the humanities. The author notes that educators are responsible for preparing students to succeed in an AI-mediated world, but this should not be limited to a pragmatic preparation for employment but should rather foster a pedagogy of care, ethics, and good judgement, while retaining a sense of exploration and wonder. In Chapter 2, Lorette Jacobs and Karin McGuirk explore the changing roles of information professionals in a HE (higher education) sector increasingly using AI, noting the increasingly central importance of this role as part of a wider professional community of practice. In Chapter 3, Erna Oliver identifies the need to work more intelligently and responsibly with AI to foster creative innovation.

In Chapter 4, Johannes Cronjé explores an integrated, process-based framework for writing with AI, but finally suggests that from a posthuman perspective, the focus should not be on the student but on the task. In Chapter 5, Karen Ferreira-Meyers examines the potential role of AI in the instructional design process. The author observes that while AI presents an immense potential for transforming HE, carefully considering ethical, social, and pedagogical dimensions is essential. The author argues the need to foreground cooperation, wisdom, and responsibility as educational values no less important than efficiency or personalisation, and the need to harmonise automation with our deepest humanity. In Chapter 6, Lilia Cheniti-Belcadhi, Mohamed Mitwally, and Asma Hadyaoui explore intelligent frameworks for assessment in AI-enhanced learning environments. They observe that while AI has the potential to greatly improve the accuracy, customisation, and efficiency of assessments, leading to a more adaptable, inclusive, and effective era in education, there are still several concerns that need to be addressed. For example, they suggest that AI assimilation and the processing of educational data pose major dangers to the privacy and integrity of such sensitive information, making student data privacy a critical issue. The authors further suggest that an effective resolution of ethical concerns requires collaboration among professionals from diverse fields such as technology, education, ethics, and policymaking.

In Chapter 7, Elmarie Kritzinger and Sarah Jane Johnston present a multi-dimensional approach to a digital transformation process in HE. They explore the interplay between key stakeholders, AI drivers, and a systematic process of commitment, strategy, testing, implementation, and evaluation. In Chapter 8, Hiro Saito argues the need to look beyond the economic and work-related potential of AI to consider how AI might be utilised to augment humanity itself in service of greater equity, solidarity, freedom, and inclusiveness, and to accelerate learning and flourishing toward collective wellbeing.

In Chapter 9, Joseph Evans Agolla and Phineas Sebopelo observe that many claims currently being made about the potential of AI are not grounded in empirical evidence and that in the African context, it is necessary to find a balance between various competing interests concerning other developments and the implementation of AI tools in education systems. In Chapter 10, in a related discussion, Phineas Sebopelo and Joseph Evans Agolla then argue that AI technology will generate opportunities for accelerating transformation in education. ODL institutions should therefore strategise on how to leverage AI and new technologies to enhance efficiency, encourage learning, and foster creativity while ensuring fairness and equity. They suggest using AI to reshape students’ experiences, enhance assessment, streamline administrative tasks, and personalise learning.

In Volume 2, Chapter 1, Samuel Amponsah, Micheal van Wyk, and Michael Adarkwah call for a more holistic and adaptive approach to academic integrity in the digital age. They suggest that integrating GPT detectors and technological tools should be complemented by a commitment to addressing the root causes of academic anxiety. By fostering a culture of trust, transparency, and resilience, they suggest that we can navigate these challenges successfully, ensuring that the pursuit of knowledge remains a transformative and enriching experience for students.

In Chapter 2, Nicky Tjano argues that there is a need to create an educational future that is both technologically advanced and morally sound. This involves tackling issues presented by concern about mounting students’ dependency on AI, reduced human interaction, communication, and ethical gaps. The author suggests that the journey to maintaining academic integrity in the AI age requires ongoing self-examination, adjustment, and a resolute dedication to the values that constitute the core of intellectual endeavours.

In Chapter 3, Faiza Gani explores the link between AI and QA (quality assurance) in HE. The author notes that the opportunities presented by AI link to QA factors and can improve these factors. On the other hand, there is also a link between the challenges provided and QA factors. Consequently, for QA factors not to be compromised, the challenges presented by AI must be carefully mitigated.

In Chapter 4, Brenda van Wyk and Marlene Holmner trace developments in information ethics through history. They suggest that we are currently at a critical point where the swift progress in AI technology necessitates a proactive and deliberate approach to ethical considerations. They also suggest that the OECD (Organisation for Economic Cooperation and Development) FIP (fair information practice) standards serve as a fundamental framework for safeguarding data and ensuring privacy, which are of utmost importance in the era of AI. These principles, in conjunction with the guidelines for responsible AI, they argue, provide a clear path for the ethical, transparent, fair, and inclusive development of AI.

In Chapter 5, Nicky Tjano unpacks the role of big data, AI, and predictive analytics in education, suggesting that creating hybrid mechanisms for review processes will assist in mitigating risks. He further adds that this can be augmented by introducing new specialised committees to ensure that unfamiliar or too technical research projects related to AI and big data are properly assessed for ethical compliance. These changes mean that educators, research review committees, and educational institutions need to relook at their governance processes, protocols, and policy frameworks.

In Chapter 6, Geesje van den Berg suggests that as AI tools will only get better and more human-like, so HE will have to adapt accordingly. Although Gen-AI has the potential to improve teaching and learning for everyone, the key is finding a balance and using Gen-AI to enhance, rather than replace, traditional educational practices. The author argues that collaboration among lecturers, researchers, policymakers, and students will then be essential to ensure that AI is used ethically and responsibly in education. The goal is to create a more equitable and effective education system that provides students with personalised teaching, feedback, and support by addressing the challenges posed by AI technologies and leveraging their benefits.

In Chapter 7, Geesje van den Berg and Patience Mudau explore the potential benefits and challenges that AI presents from the perspective of educators. They note that from their research, the most prominent benefits identified were that these digital tools can provide individualised, 24/7 guidance and support, and also augment learning content. They also note that these tools can assist with simplifying language, a unique benefit in multilingual contexts. Furthermore, they argue that Gen-AI tools provide opportunities for active involvement and that they can save time and make information and knowledge accessible. However, they note concerns that the technologies could provide biased information, lead to decreased human interaction and lack emotional intelligence. Additionally, it could provide inaccurate information and has a risk that students can over-rely on these tools. They have further identified the risks of plagiarism and data privacy and security as challenges. These guidelines focus on the need for policies and AI literacy training, setting an example and providing guidelines to students using Gen-AI. Another important point is that the Gen-AI tools should complement, and not replace, the human elements of teaching and learning, ensuring a holistic and inclusive educational approach.

In Chapter 8, Elize du Plessis argues the need to reimagine online assessment methods observing that Gen-AI holds immense potential to revolutionise how we evaluate knowledge and skills. From adaptive assessments to automated grading and feedback, these models can enhance student and educator assessment experiences. However, careful attention must be given to addressing challenges such as bias, reliability, security, and privacy. Collaboration between researchers, educators, and policymakers is necessary to leverage the benefits of LLMs while upholding ethical standards and ensuring the fairness and validity of online assessments. She further argues that academics should investigate Gen-AI, identify its limitations, evaluate its potential applications within the relevant disciplines or teaching contexts, and engage in discussions about all these aspects with students, who are likely already familiar with the technology.

In Chapter 9, Micheal van Wyk explores the issue of academic dishonesty and how it might be prevented. The author suggests that academics must redesign authentic context-based assessment tasks, such as specific case studies, problem-solving activities, project-based learning tasks, and academic writing skills. The research report suggests that education faculties need to update AI policies, provide guidelines, and educate students about AI literacy skills and AI detector tools for the ethical use of Gen-AI tools in teaching and learning. The author argues that this must include developing students’ ability to use AI in appropriate ways.

The nineteen chapters in this two-volume publication offer insights into both challenges and opportunities for working with AI in general and Gen-AI in particular. Overall, the argument that emerges is for a reasoned engagement with the evolving possibilities of the new technology, but as Mark Twain is credited with observing, gradual improvement is probably going to be more useful than delayed perfection.

References

Abulibdeh, A., Zaidan, E., & Abulibdeh, R. 2024. Navigating the

confluence of artificial intelligence and education for

sustainable development in the era of industry 4.0: Challenges, opportunities, and ethical dimensions. Journal of Cleaner Production 437(1). 140527. https://doi.org/10.1016/j. jclepro.2023.140527

Baidoo-Anu, D. & Owusu Ansah, L. 2023. Education in the era of generative artificial intelligence (AI): Understanding the potential benefits of ChatGPT in promoting teaching and learning. Journal of AI 7(1):52-62. https://doi.org/10.2139/ ssrn.4337484

Bates. AW. n.d. 12. Using SECTIONS to select digital tools. Pressbooks. Available at: https://pressbooks.pub/everydayid/chapter/using sections-to-select-digital-tools/ . (Accessed on 13 December 2024).

Bates. AW. 2022. Teaching in a digital age: Guidelines for designing teaching and learning. 3rd ed. Available at: https://pressbooks.bccampus. ca/teachinginadigitalagev3m/ . (Accessed on 11 December 2024).

Bates, AW. 2024. What should universities do about AI for teaching and learning? Online Learning and Distance Education Resources. 26 April 2024. Available at: https://www.tonybates. ca/2024/04/26/what-should-universities-do-about-ai-for teaching-and-learning/ . (Accessed on 10 December 2024).

Bozkurt, A. & Sharma, RC. 2023. Generative AI and prompt engineering: The art of whispering to let the genie out of the algorithmic world. Asian Journal of Distance Education 18(2):i-vi.

Carrington, A. 2015. The padagogy wheel: It’s not about the apps, it’s about the pedagogy. Available at: https://www.teachthought. com/technology/the-padagogy-wheel/ . (Accessed on 1 December 2024).

Carrington, A. 2024. The padagogy wheel, ENG AI V7.0. Available at: https://designingoutcomes.com/assets/PadWheelV7/PW_ENG_ AI_V7.0.pdf . (Accessed on 1 December 2024).

COL (Commonwealth of Learning). 2023. Samoa pioneers AI-powered learner support. Commonwealth of Learning. 21 August 2023. Available at: https://www.col.org/news/samoa-pioneers-ai powered-learner-support/ . (Accessed on 14 December 2024).

COL (Commonwealth of Learning). 2024. USP enhanced its Semester Zero programme with GPT-powered AI support. Commonwealth of Learning . 22 April 2024. Available at: https://www.col.org/ news/usp-enhanced-its-semester-zero-programme-with gpt-powered-ai-support/ . (Accessed on 14 December 2024).

COL (Commonwealth of Learning) & UNESCO (United Nations

Educational, Scientific, and Cultural Organisation). 2024. UNESCO. Global education monitoring report 2024. Pacific: Technology in education: A tool on whose terms? Available

at: https://unesdoc.unesco.org/ark:/48223/pf0000391211 . (Accessed on 2 December 2024).

Cooper, G. 2023. Examining science education in ChatGPT: An exploratory study of generative artificial generative intelligence. Journal of Science Education and Technology 32:444 452. https://doi.org/10.1007/s10956-023-10039-y

Denny, P., Gulwani, S., Heffernan, NT., Käser, T., Moore, S., Rafferty, AN., & Singla, A. 2024. Generative AI for Education (GAIED): Advances, opportunities, and challenges. Arxiv. 10 pages. Available at: https://arxiv.org/pdf/2402.01580 . (Accessed on 5 December 2024).

Du, H., Sun, Y., Jiang, H., Atiquil Islam, AYM., & Gu, X. 2024. Exploring the effects of AI literacy in teacher learning: An empirical study. Humanities and Social Sciences Communications 11. Article 559. 10 pages. https://doi.org/10.1057/s41599-024-03101-6

Eke, DO. 2023. ChatGPT and the rise of generative AI: Threat to

academic integrity? Journal of Responsible Technology 13. 100060. 4 pages. https://doi.org/10.1016/j.jrt.2023.100060

Hegarty, B. 2015. Attributes of open pedagogy: A model for using open educational resources. Educational Technology July-August:3-13.

Kizilcec, RF. 2024. To advance AI use in education, focus on understanding educators. International Journal of Artificial Intelligence in Education 34:12-19. https://doi.org/10.1007/ s40593-023-00351-4

Law, L. 2024. Application of generative artificial intelligence (GenAI) in language teaching and learning: A scoping literature review. Computers and Education Open 6. 13 pages. https://doi. org/10.1016/j.caeo.2024.100174

Linderoth, C., Hultén, M., & Stenliden, L. 2024. Competing visions of artificial intelligence in education – a heuristic analysis on sociotechnical imaginaries and problematizations in policy guidelines. Policy Futures in Education 22(8):1662-1678. https:// doi.org/10.1177/14782103241228900

Mays, T. 2024a. Rooting for robots through pragmatic pedagogy. Keynote address to the UNISA International Conference on Teaching and Learning, Pretoria, South Africa on 7 May 2024. Available at: http://hdl.handle.net/11599/5599. (Accessed on 12 December 2024).

Mays, T. 2024b. Pedagogy 4.0: Technology enhanced teaching and learning. Keynote address to the Teaching with 21st Century Technology Conference at the University of the West Indies, Moana, Jamaica, 28 November 2024. Available at: http://hdl. handle.net/11599/5587 . (Accessed on 12 December 2024).

Millwood, R. 2021. Learning theory. HoTEL (holistic approach to technology enhanced learning). Available at: https:// cmapspublic3.ihmc.us/rid=1WRDG1P8T-5LW948-1MKQ . (Accessed on 6 December 2024).

Mohamed, A. & Mishra, S. 2024. Developing policy guidelines for artificial intelligence in post-secondary institutions. COL (Commonwealth of Learning). Available at: http://hdl.handle. net/11599/5615 . (Accessed on 6 December 2024).

Nikolic, S., Daniel, S., Haque, R., Belkina, M., Hassan, GM., & Grundy,

S. 2023. ChatGPT versus engineering education assessment:

A multidisciplinary and multi-institutional benchmarking

and analysis of this generative artificial intelligence tool to

investigate assessment integrity. European Journal of Engineering Education 48(4):559-614. https://doi.org/10.1080/03043797.20 23.2213169

OER (open educational resources) Africa. 2023. Three ways artificial intelligence could change how we use open educational resources. Blog. Available at: https://www.oerafrica.org/ content/three-ways-artificial-intelligence-could-change how-we-use-open-educational-resources . (Accessed on 7 December 2024).

Onesi-Ozigagun, O., Ololade, YJ., Eyo-Udo, NL., & Ogundipe, DO.

2024. Revolutionizing education through AI: A comprehensive

review of enhancing learning experiences. International Journal of Applied Research in Social Sciences 6(4):589-607. https://doi. org/10.51594/ijarss.v6i4.1011

Oxford Languages. 2024. Artificial intelligence – definition. Available at: https://languages.oup.com/google-dictionary-en . (Accessed on 7 December 2024).

Paskevicius, M. 2024. Policy and practice of artificial intelligence in teaching and learning at post-secondary educational institutions in the Commonwealth. COL (Commonwealth of Learning). Available at: http://hdl.handle.net/11599/5605. (Accessed on 2 December 2024).

Perrault, R. & Clark, J. 2024. Artificial Intelligence Index Report 2024.

Stanford University: Human-Centered Artificial Intelligence. Ratten, V. & Jones, P. 2023. Generative artificial intelligence (ChatGPT):

Implications for management educators. The International Journal of Management Education 21(3). 7 pages. https://doi. org/10.1016/j.ijme.2023.100857

Rowntree, D. 1987. Assessing students: How shall we know them? 2nd ed. London: Kogan.

Scott, P. 2024. AI in open education. Virtual keynote address on 24 September 2024 at University of Ghana Conference on the theme ‘Embracing the future: The emergence of AI and its impact on open distance and e-learning.’ COL (Commonwealth of Learning). Available at: http://hdl.handle.net/11599/5663. (Accessed on 23 December 2024).

Smolansky, A., Cram, A., Raduescu, C., Zeivots, S., Huber, E., & Kizilcec, RF. 2023. Educator and student perspectives on the impact of generative AI on assessments in higher education. L@S ’23: Proceedings of the Tenth ACM Conference on Learning @ Scale, July 2023. https://doi.org/10.1145/3573051.3596191

Sevnarayan, K. & Potter, M-A. 2024. Generative artificial intelligence in distance education: Transformations, challenges, and impact on academic integrity and student voice. Journal of Applied Learning & Teaching 7(1):104-114. https://doi.org/10.37074/ jalt.2024.7.1.41

Sternberg, RJ. 2024. Human intelligence. Britannica.com . Available at: https://www.britannica.com/science/human-intelligence psychology . (Accessed on 17 December 2024).

Stoker, M. 2024. Collaborating with AI to create a collaborative classroom. Presentation CCBYSA, Personal Communication. YouTube . Available at: https://www.youtube.com/ watch?app=desktop&v=Fbr6RMOxyt0 . (Accessed on 12 December 2024).

Williamson, B. 2024. The social life of AI in education. International Journal of Artificial Intelligence in Education 34:97-104. https:// doi.org/10.1007/s40593-023-00342-5

Introduction

Academic integrity is a fundamental principle of education and serves as the bedrock upon which the integrity of knowledge and scholarship rests (Bretag, Harper, Burton, Ellis, Newton, Rozenberg, Saddiqui, & Van Haeringen 2019:1849). In today’s digital age, where the boundless expanse of information is merely a click away and educational landscapes have gravitated towards online platforms, the challenges to maintaining academic integrity have metamorphosed (Eaton & Gysbers 2021:48). As asserted by Kumar (2020:133), students’ encounters with academic anxiety have surged, complicating the educational landscape even further.

Against this backdrop, the emergence of GPT (generative pre-trained transformer) detectors beckons both promise and disquiet among major stakeholders in education, especially the academy. These advanced AI (artificial intelligence) tools have been meticulously designed to unearth and thwart plagiarism and cheating. They also have the potential to reconfigure the very fabric of academic integrity which has aroused questions regarding the place of critical thinking among students and the current generation. However, the debate on academic malfeasance seems to have no end in sight from the human perspective, as GPT detectors are also casting shadows of privacy infringement, fairness disparities, and the omnipresent question of technology’s role in modern education (Jones & Lee 2023:15).

By way of definition, academic integrity embodies the pledge to honesty, originality, and ethical conduct in all facets of the learning journey (Bretag et al. 2019:1851). Safeguarding academic integrity upholds a culture of trust and respect within the academic community, ensuring that the rewards – grades and degrees – echo genuine understanding and competence. Nevertheless, the digital era has ushered in a new era for academic integrity, one riddled with pitfalls and challenges (Eaton & Gysbers 2021:47). In addition, the internet’s vast sea of knowledge and information at the disposal of students presents them with the burgeoning temptation to plagiarise or engage in acts of dishonesty at every point in time. In response to these evolving circumstances, IHEs (institutions of higher education) have found themselves at a crossroads over the past few years. At the same time, this has compelled many IHEs to devise innovative strategies to preserve the sanctity of learning and assessments. According to Steele (2023:5), the internet is filled with falsehood, fake information, and no shortage of misinformation.

We concur that in a world where everything can be faked, it is practical to identify real truth. Important questions such as the following beg answers: ‘Who did quality assurance on programmes or products?’ and ‘Who has done the quality assurance (checks and balances) of the content generated in the tech companies?’ Moreover, several studies have reported the dangers of chatbots (Eaton & Gysbers 2012:57; Liang, Yuksekgonul, Mao, Wu, & Zou 2023:1 of 9). Unless prompted otherwise, GPT detectors can potentially threaten entire academic projects. Moreover, while the digital age has revolutionised how we access and disseminate information, it has engendered a pandemic of academic anxiety among students (Kumar 2020:139). Contemporary students find themselves caught in the crossfire of mounting academic pressure and elevated expectations for success. The relentless competition facilitated by digital platforms and the omnipresence of educational metrics magnify this pressure.

As students grapple with the looming spectre of failure or falling short of academic standards, stress and anxiety become their companions. Regrettably, some resort to unethical practices to stave off these anxieties and safeguard their academic standing. This dates back to the days when students would plagiarise large doses of materials into their projects or assignments without due recourse to the original owners of the work or the consequences when they were caught out engaging in such unethical behaviours. It is, however, worth noting that academic worries, when left unchecked, can exert a profound toll on students’ mental wellbeing and their capacity to engage in critical thinking processes that define genuine learning experiences and authentic assessments.

In response to the multifaceted challenges presented by the digital age, a collaborative alliance between IHEs and technology companies has given birth to innovative solutions (Smith & Smith 2022:178). Among these solutions, GPT detectors, powered by the advanced capabilities of natural language processing and machine learning algorithms have been developed to address issues of academic misconduct. Their primary mission lies in the identification of instances of plagiarism, cheating, and other forms of academic dishonesty (Jones & Lee 2023:17). By meticulously scrutinising written assignments, essays, and academic works, they search for any signs of unoriginal content or improperly cited sources. Through the harnessing of AI, GPT detectors are believed to furnish a formidable defence against academic misconduct, serving as the vanguards of educational sanctity.

Considering the preceding, GPT detectors undertake a comprehensive examination by comparing the work of students with extensive databases of academic and non-academic content (Smith & Smith 2022:179). Their keen algorithms detect similarities and anomalies that may suggest plagiarism or unauthorised collaboration. Upon the discovery of such discrepancies, the detectors swiftly alert instructors or academic integrity committees, who in turn, are empowered to take the requisite measures (Jones & Lee 2023:19). By automating this process, GPT detectors unburden educators from the onerous task of policing academic dishonesty, thus granting academics the ability to channel their energies into the noble pursuit of teaching and nurturing critical thinking among their students.

Despite the promise they hold, GPT detectors introduce a trove of ethical and privacy concerns (Bretag et al. 2019:1843). The implementation of these technologies invariably necessitates the intrusive surveillance of students’ academic work and online activities (Eaton & Gysbers 2021:59). There have been instances where GPTs have relayed detected plagiarism in students’ work which proved otherwise upon further probe. The lack of understanding of the usage of such tools among educators also cultivates erroneous judgements on students’ work at times. It is for such reasons that some critics argue that this heightened surveillance infringes upon students’ privacy rights and has the potential to stifle open communication and free expression in educational settings (Kumar 2020:134). As a result, striking the delicate balance between preserving academic integrity and fostering an environment conducive to creativity and innovation poses an intricate challenge for IHEs.

Furthermore, the fairness of GPT detectors raises apprehensions (Smith & Smith 2022:188). These systems, although intelligent, may inadvertently penalise students whose writing styles bear resemblance to existing academic or online content, even when the students’ work is genuinely original. The spectre of false positives and negatives looms large, potentially inflicting unjust academic penalties and jeopardising the academic journey of students with no ill intent (Jones & Lee 2023:21). Again, this brings to the fore the need for educators to have in-depth knowledge of how these GPT detectors function and the skill to probe the systems in order not to unfairly penalise students for their hard work.

While genuine concerns exist regarding the use of GPT detectors, these tools also bear the potential to become catalysts for the development of critical thinking skills in students (Eaton & Gysbers 2021:56). By acting as deterrents against academic dishonesty, GPT detectors sculpt an environment where students are not merely encouraged but compelled to engage in critical and independent thinking. They understand that their work will be assessed on its intrinsic value rather than its originality, thereby ushering in a new era of authentic and enriched learning experiences (Kumar 2020:140). Thus, there is a need to comprehensively explore the broader implications of GPTs on critical thinking in the educational realms.

In the context of the foregoing, this chapter argues that academic anxiety, technostress, and integrity related to the use of GPTs exist and theorises plausible resolutions to remedy these situations among students. To achieve this, the ensuing sections provide an overview of GPT detectors, delves deeply into the concerns raised with academic integrity and anxiety, and charts the way out of this situation.

Overview of GPT Detectors

In the age of digital learning and the ubiquitous availability of information, the need to maintain academic integrity has never been more crucial. The evolution of technology and educational methods has paved the way for innovative tools designed to detect and prevent academic dishonesty. Among these, GPT detectors stand out as sophisticated AI solutions specifically engineered to identify and deter plagiarism, cheating, and other forms of academic misconduct (Smith & Smith 2022:176). Amongst the most well-known GPT detectors are GPTZero, ZeroGPT, GPT 2 output detector, CheckGPT, and Writefull. According to Steele (2023:3), these detectors can be used to enhance students’ critical thinking, free up their intellectual passion and reading comprehension, and teach academic writing skills. Given the preceding, there is a need to provide a comprehensive overview of GPT detectors, delving into their development, functionality, and their role in reshaping academic integrity in the modern educational landscape.

GPT detectors are products of the growing intersection between AI and education. They have evolved in response to the escalating challenges posed by the digital age, where vast repositories of information are easily accessible, and online learning platforms have become the norm (Jones & Lee 2023:23). These detectors are derivatives of the broader GPT family of language models, which include models such as GPT 3, GPT 4, and beyond. GPT language models are pre-trained on massive datasets that encompass a wide array of texts, ranging from books and academic articles to internet content. This pre-training equips them with an extraordinary grasp of natural language understanding and generation.

GPT detectors, however, are specialised variations of these models, fine-tuned to tackle specific issues including academic integrity and anxiety. The fine-tuning process involves training the model on a dataset of academic content, including essays, research papers, and various academic resources (Smith & Smith 2022:186). This specific training enables GPT detectors to distinguish between authentic academic work and content that may be plagiarised or in violation of academic integrity. GPT detectors aim to maintain and promote academic honesty. Bretag et al. (2019:1855) summarise the functionality of the chatbots performing the following sequential activities: Content analysis, data comparison, similarity assessment, citation, referencing analysis, alerting, and reporting. With the overview of GPT detectors provided, the critical issues of academic anxiety, technostress, and academic integrity are discussed below.

Analysing the Landscape of Academic Anxiety and Technostress

Academic anxiety is emotionally exhausting and involves feelings of worry, distress, and tension regarding the potential failure of an assignment or exam. A student’s personal life and academic performance could be affected as well. A correlation study by Shakir (2014:31) reports a negative correlation between academic anxiety and academic performance among secondary school students. Additionally, Das (2019:20) conducted a similar study with mathematics students and discovered that academic anxiety is significantly negatively related to students’ academic performance. Thus, the concentration and motivation of students are disrupted by high levels of anxiety. Moreover, academic anxiety, fuelled by technostress, became a widely discussed topic during and after the COVID-19 (Coronavirus disease of 2019) pandemic and the hype surrounding generative AI tools like ChatGPT 3.5 (chat generative pre-trained transformer 3.5) (Amponsah, Van Wyk, & Kolugu 2022:7). Technostress itself is defined as the negative psychological (e.g., anxiety), physiological (e.g., fatigue), and behavioural (e.g., exhaustion) responses humans have to technology overload (Kumar 2024:1 of 16). If not properly managed by IHEs, academic anxiety induced by technostress can result in a cognitive avoidance of distress, uncertainty, and psychological discomfort.

It is imperative to add that the past four decades of research on academic anxiety has predominantly been characterised by cognitions and behavioural patterns of students which have remained consistent as dimensions of poor academic performance. Also, scholars have reported patterns of academic anxiety which are associated with fear of negative evaluations (Brown, White, Doan, & De Bruin 2011:331), poor study skills (Sanghvi 1995:71), poor test performance (Kivimäki 1995:51), perfectionism (Eum & Rice 2011:167), and recently, technostress (Amponsah, Van Wyk, & Kolugu 2022:9; Huo & Siau 2023:2). Furthermore, as there has been an increase in the development of Gen-AI (generative artificial intelligence) tools like ChatGPT

3.5 and 4.0, so has academic anxiety and technostress increased amongst students which have affected their concentration, left them with distracted attention, excessive anxiety, tension, fear of having delays and avoidance of assignments. This also affected their general academic achievement and wellbeing (Caporusso 2023:3 of 12).

Studies on Gen-AI, like ChatGPT 3.5 and Google Bard as LLMs (large language modules) have disrupted and significantly impacted education positively. More importantly, it has the potential to transform and increase accessibility and accelerate education exponentially. In the chase for academic success, there are downsides to using chatbots. All these Gen-AI tools are available and have created a hype among students as they view these tools as easy options to answer assessments and develop projects in pursuit of instant success. Based on the exposure to LLMs such as ChatGPT 3.5, research has reported concerns about using ChatGPT to plagiarise or cheat in their assignments (un) knowingly because of mounting academic pressure and elevated expectations for success (Cotton, Cotton, & Shipway 2023:236).

However, this prompt engineering phenomenon has some major drawbacks related to technostress, academic anxiety, and academic dishonesty (Ayyagari, Grover, & Purvis 2011; Huo & Siau 2023:2; Wach, Duong, Ejdys, Kazlauskaitė, Korzynski, Mazurek, Paliszkiewicz, & Ziemba 2023:19). Additionally, the overuse of Gen-AI could increase academic anxiety and harm a student’s academic performance if not correctly and mildly used for academic purposes. Studies indicate that students have plagiarised their assignments by using Gen-AI tools to cheat in academic writing which is an academic violation and dishonesty (Azulay Chertok, Barnes, & Gilleland 2014:1324; Caporusso 2023:3 of 12) which often leads to an increase in academic anxiety. Academic anxiety is regarded as a psychological and emotional statement of the mind which leads to feelings of stress, tension, low concentration levels, and fear of failure in an academic task. These distractors affect an individual’s self-esteem, motivation, and avoidance of submitting assignments timeously, which ultimately affects their academic performance.

Moreover, the literature reports technostress and technooverload as downside effects of social media and ChatGPT. In 1984, the concept of technostress was first used by clinical psychologist, Craig Brod, as a major cause of the overuse of technology by people. According to Amponsah et al. (2022:3), excessive use or exposure to social media can lead to technostress, cognitive overload, reduced staff morale, low productivity, and ‘academic fatigue.’ Additionally, studies have indicated that social media can lead to technostress and techno-overload (when users are compelled to use technologies faster and longer) (Leung & Zhang 2017:389). However, Gen-AI has recently brought significant changes and transformations to HE (higher education), but it has also led to academic dishonesty and technostress among students (Ragu-Nathan, Tarafdar, Ragu-Nathan, & Qiang 2008:423; Khanthavit & Khanthavit 2023:213). Gen-AI tools such as ChatGPT 3.5, LlAMA 2, and Bard are being used more by students for academic purposes, but excessive exposure could lead to increased technostress and techno-anxiety (Chatzopoulou, Filieri, & Dogruyol 2020:1272-1273).

In a critical analysis study, Wach et al. (2023:9) explored the ‘dark side’ of the overuse of prompt engineering like ChatGPT. The authors found that drawbacks such as threats and anxiety are major causes for concern relating to the use of ChatGPT and provided recommendations to prevent academic anxiety. This is in addition to findings that technostress is experienced by people who have excessive exposure to and overuse in reliance on technology (techno-overload) or ICTs (information and communication technologies) (Singh, Bala, Dey, & Filieri 2022:271).

Technostress results from excessive screentime and the overuse of technology either at school, work, or in private life. This has several negative consequences such as poor work performance, depression, job dissatisfaction, and reduced work commitment (Qi 2019:1338). For example, when students face academic pressure and expectations for success, the only way out for some of them is the excessive use of chatbots as a means to cheat academically. In addition, technostress is experienced by people who have excessive exposure to and overuse in reliance on technology (techno-overload) or prolonged ICT exposure in the workplace (Singh et al. 2022:271). The cause of excessive use of ChatGPT is academic anxiety which is the inability to cope with the academic pressure when chasing success. Moreover, recent studies reported the downside of prolonged exposure to and the use of chatbots (prompt engineering). In this case ChatGPT is a cause of technostress and fatigue among students and staff (Wach et al. 2023:9). This view is supported by Caporusso (2023:9 of 12) with reference to creative displacement anxiety and technostress. Moreover, Caporusso argues that IHEs need to increase the awareness of the overuse of technology for the wellbeing of staff.

Concerns with Academic Integrity

Cheating, plagiarism, and other dishonest acts, which are violations of academic integrity, are an increasing concern for faculty, students, and the public who hold the trust in graduates to possess the knowledge needed to complete their studies (Azulay Chertok et al. 2014:1324). As already highlighted, common examples of academic dishonesty cited include cheating in examinations and plagiarising written assignments (Turner & Beemsterboer 2003:1122). Mostofa, Tabassum, and Ahmed (2021:257) document that academic plagiarism has been on the rise globally and is commonly widespread among university students. Since the introduction of LLMs such as ChatGPT and Google Bard, many scholars have raised ethical concerns regarding academic integrity. That is, with the transformative potential of the modern and pervasive technological revolution of AI chatbots, concerns about academic integrity have become more critical than ever before (Bin-Nashwan, Sadallah, & Bouteraa 2023:2 of 11).

The increased comfort with AI technologies makes it easier for students in online learning environments to engage in violations of academic integrity such as copy-pasting from online sources and the unauthorised use of electronic resources during exams (Azulay Chertok et al. 2014:1325). This exponential growth of unethical behaviour in HE poses crucial challenges to quality education as enshrined in SDG 4 (Sustainable Development Goal 4) and the tenets of academic integrity (Gottardello & Karabag 2022:527). As a result, universities are going through an unprecedented disruption and concerns relating to the breach of academic integrity in light of the social, economic, and technological changes in HE (Roe & Perkins 2022:1 of 10).

Furthermore, plagiarism detection software such as Turnitin and SafeAssign, as well as the GPT detectors already indicated above are designed to combat outsourcing assessment and research tasks (Liu, Yao, Li, & Luo 2023:6 of 23; Perkins, Roe, Postma, McGaughran, & Hickerson 2023:4 of 21), although with new LLMs such as ChatGPT, it is difficult for inexperienced faculty members and researchers to detect GPT-generated texts (Liu et al. 2023:5 of 23). However, thanks to these tools, plagiarism is being discovered at a faster rate than in the past (Halupa & Bolliger 2013:298). Many IHEs, therefore, employ plagiarism detection tools to promote novelty in writing and detect novel or unexpected ways of cheating or plagiarising (Canzonetta 2021:1).

According to Belli, Raventós, and Guarda (2020:661), many students plagiarise because they lack confidence in their writing abilities, do not dedicate sufficient time to accomplish tasks, possess a positive attitude towards deception, or simply ignore how to properly quote a text. Ayton, Hillman, Hatzikiriakidis, Tsindos, Sadasivan, Maloney, Bragge, Diug, and Illic (2022:1925) have also found that a limited understanding of plagiarism and poor academic or language skills are among the reasons why students plagiarise. Belli et al. (2020:661) add that the cost of plagiarism is very expensive because it turns educators into police officers who have to dedicate much of their time and effort to something that does not yield any benefit to the learning environment. This invariably affects the performance of educators as they spend precious time and effort to police students’ work at the expense of research and other fulfilling engagements.

Nonetheless, in many IHEs, both faculty and students do not understand the concept of plagiarism, especially self-plagiarism (Halupa & Bolliger 2013:303). Mostofa et al. (2021:258) assert that many students in different disciplines and academic levels have a poor awareness of plagiarism. For example, in a report published on Unlocking the power of generative AI models and systems such as GPT 4 and ChatGPT for higher education, some of the essential questions that students asked are, ‘Am I allowed to use ChatGPT for a seminar or final paper, or is that cheating?; and ‘How exactly do I use ChatGPT best?’ (Gimpel, Hall, Decker, Eymann, Lämmermann, Mädche, Röglinger, Ruiner, Schoch, Schoop, Urbach, & Vandirk 2023:3).

According to Mostofa et al. (2021:265), not having clear ethical norms and standards about academic writing can cause an individual to fall foul of the guilt of plagiarism. While the media coverage about ChatGPT and recent LLMs has focused on ‘cheating or plagiarism’ (Gimpel et al. 2023), Perkins (2023:1 of 24) argues that the use of LLMs does not necessarily indicate plagiarism if students are transparent in how they have been used in any submission. There is also intentional and unintentional plagiarism (Ayton et al. 2022:1929), implying that it is possible for a student to unintentionally fall victim to academic misconduct or the breach of (institutional) plagiarism rules because of the nature (sensitivity) of plagiarism detection tools.

To conclude, Turner and Beemsterboer (2003:1123) acknowledge that academic dishonesty harms the dishonest individual because they do not acquire the basic learning, skills, or knowledge. Subsequently, the public who holds trust in the graduate’s skill is deceived and the peer of the dishonest individual who competes in class suffers harm due to an unfair advantage. The faculty of the dishonest student is harmed by not being aware of the true grasp of the content. Hence, such faculty are unable to modify instruction when necessary. Finally, the reputation of the university or institution of the dishonest student is often badly damaged when the public perceives that it does not set and uphold higher standards among both faculty and students.

Given the preceding, it is necessary to identify that academic integrity demands a student to adopt the practice of engaging in meaningful research and completing academic work in a fair and coherent manner (Bin-Nashwan et al. 2023:4 of 11). Bin-Nashwan et al. emphasise that fostering academic integrity requires dedication, honesty, trust, responsibility, and respect. Similarly, the International Centre for Academic Integrity (2021) has conceptualised the fostering of academic integrity as a commitment to six basic principles including honesty, trust, fairness, respect, responsibility, and courage. Therefore, IHEs are required to develop student competencies and engage in joint action with all academic communities to preserve academic integrity (Gottardello & Karabag 2022:527-529).

Given the above, we argue that academic anxiety has negative consequences. Still, there is a need to foster integrity in the age of GPT detectors to address misinformation and the numerous falsehoods generated by Gen-AI tools. This leads to the question, ‘Can Gen-AI plagiarism tools like GPT detectors be utilised to promote critical thinking instead of detecting dishonesty and cheating among students and academia?’ To preserve the image and reputation of IHEs, it is essential to implement targeted measures against cheating. To protect academic integrity, it is argued that awareness and ethical considerations are crucial, along with possible revisions to policies related to integrity, copyright infringements, and academic practices. The latter provides a basis for IHEs to implement anti-cheating strategies.

The literature offers measures to prevent academic cheating by Gen-AI tools in education. For example, Oravec (2023:220

223) has conducted studies on cheating detection strategies and Gen-AI empowered skills to aid students in addressing academic dishonesty and plagiarism. One of their major recommendations was for educators to promote and raise awareness about the educational benefits of using Gen-AI, specifically GPT detectors, to enhance students’ and educators’ proficiencies in utilising Gen-AI tools for preventing dishonesty. Another measure to prevent cheating or academic dishonesty is using contextual case studies, providing original research and problem-based projects for students to present the results, writing, and reporting as a group. Other recommended approaches include group discussions, the development of purpose-driven objectives like specific case studies, problem-solving activities, project-based learning tasks, and empowering students with academic writing skills. We strongly believe that implementing these strategies will raise awareness and improve critical conversations about Gen-AI tools, safeguarding academic integrity and fostering students’ sense of integrity.

Since the emergence of GPT detectors, universities have been compelled to address policy revisions, while advocating awareness initiatives among staff and students is of vital importance to prevent dishonesty. Universities need to create an awareness of and view ethical considerations for using Gen-AI tools or GPT detectors as important considerations before adopting chatbots in practice. As a matter of urgency, academics must be cognisant of the speed of the emergence of new Gen-AI tools such as ChatGPT 3.5 and other LLMs. Studies concur that a major issue that needs to be addressed is the ethics of using Gen-AI and detector tools by students and academics (Hagendorff 2020:109). These studies express concerns of fear that breach ethical implications. Among similar views expressed related to the challenge of cheating, plagiarism, and copyright infringement all have ethical implications for the quality of education.

Furthermore, combating academic dishonesty requires a multifaceted approach which might include tutorials and other short educational interventions aimed at increasing the knowledge of faculty or students about academic integrity (Stoesz & Los 2019:3). Azulay Chertok et al. (2014:1325) enumerate several means to foster academic integrity among students including consistently meting out disciplinary action for the violation of academic integrity, educating students about academic dishonest behaviours and their consequences, providing clarity in policy and expectations about academic dishonesty, setting up measures to eliminate or reduce possibilities of cheating, developing new methods of evaluation which are in accordance with the advancement of technology, implementing an integrity agreement between both faculty and students, and also the implementation of honour codes (ethical principles governing an academic community such as an agreement not to lie or cheat). In consensus, Bultas, Schmuke, Davis, and Palmer (2017:58) also mention that the development and clear enforcement of integrity policies, the role modelling of professional behaviour, and integrity by faculty can influence student behaviour as the use of honour codes are solutions to improve the academic integrity among students.

McCabe and Pavela (2004:12-15) list ten core principles of academic integrity on how faculty should foster student honesty:

It is also believed that a careful design of assessment tasks in the form of authentic assessments can deter students from engaging in academic misconduct and further promote ethical decision-making and behaviour regarding academic work (Sotiriadou, Logan, Daly, & Guest 2020).

Moreover, educators also need to be involved in the formulation and implementation of academic integrity policies because their understanding and beliefs about it are crucial for effective integration (Gottardello & Karabag 2022:527). Academic integrity is less likely to be violated when students perceive that there is a commitment to ethical standards from all sides (allacademic community) and information about it is adequately disseminated (Gottardello & Karabag 2022). Gottardello and Karabag (2022) add that involving academics and different stakeholders can contribute to forging a culture of academic integrity and identifying ways to appropriately respond to the expectations of their institutions. Also, the publication of academic integrity policies in places widely and readily accessible to students such as web pages and student manuals can facilitate its implementation. This can occur in tandem with discussion forums on academic integrity to increase awareness of the policies (Gottardello & Karabag 2022; Whitley & Keith-Spiegel 2001:332).

Lastly, institutions and educators can also utilise technologies such as Turnitin for identifying the use of APTs (automated paraphrasing tools) or LLMs for academic dishonesty (Roe & Perkins 2022:5 of 10). Additionally, both faculty and students need to be trained on the proper use of APTs and LLMs to avoid academic misconduct (Roe & Perkins 2022:7 of 10). A national study of first-year computing programmes also reveals several strategies for maintaining academic integrity such as education about academic integrity; discouraging cheating through monitoring or observing student work; reducing the benefits of cheating through low-stakes assessment; making cheating difficult by invigilating assessment; and empowerment by supporting and building relationships among students (Sheard et al. 2017:246), fostering academic integrity matters because it signifies that honesty persists in all endeavours of an academic environment (Turner & Beemsterboer 2003:1122).

Synthesis of the Critical Issues

Academic integrity, being a foundational principle in education, faces unprecedented challenges in the contemporary digital era. The ubiquitous availability of information, the prevalence of online learning platforms, and the transformative role of GPT detectors demand further comprehensive exploration of the issues. This in-depth analysis thus navigates through the complexities of academic integrity, the surge in academic anxiety, the evolving landscape of GPT detectors, and critical strategies for fostering and preserving academic probity.

First, it is obvious that academic integrity has been compromised largely with the advent of the digital age. Academic integrity is regarded as the bedrock of education and it traditionally encapsulates a commitment to honesty, originality, and ethical conduct (Bretag et al. 2019:1849). However, the advent of the digital age has redefined the parameters of maintaining academic probity. With information accessible at the click of a button and educational landscapes shifting to online platforms, IHEs grapple with new challenges in upholding the sanctity of academic integrity (Eaton & Gysbers 2021:47).

Second, accompanying the digital revolution is a palpable rise in academic anxiety among students, a phenomenon that Kumar (2020:133) asserts, has surged significantly. The relentless competition facilitated by digital platforms and the omnipresence of educational metrics amplify this anxiety. As students confront the looming spectre of failure or falling short of academic standards, stress becomes a prevalent companion, prompting some to resort to unethical practices as a coping mechanism (Singh et al. 2022:270).

Also, amidst these challenges, the emergence of GPT detectors, exemplified by models like GPTZero and CheckGPT, introduces both promise and disquiet. These advanced AI tools are meticulously designed to unearth and thwart plagiarism, cheating, and other forms of academic dishonesty, presenting a paradigm shift in how IHEs combat breaches of academic integrity (Hagendorff 2020:113). However, while GPT detectors offer a promising defense against academic misconduct, they simultaneously raise ethical and privacy concerns. The implementation of these technologies necessitates an intrusive surveillance of students’ academic work and online activities, leading to debates about potential infringements of privacy rights and the impact on open communication in educational settings (Bretag et al. 2019:1855). It is vital to expose students to detector software to discourage cheating, which may prevent dishonesty among students. Awareness should be inculcated, as well as the adoption and design of authentic learning experiences to highlight the usefulness and educational value of GPT detector software. As a strategy, webinars should be created to promote critical conversations among students and staff as a means to increase awareness of GPT detector software. We are, therefore, advocating for the need to carefully strike the delicate balance between preserving academic integrity and fostering an environment conducive to creativity and innovation.

The intricate interplay between academic anxiety, technostress, and the utilisation of Gen-AI tools also forms a critical nexus in the modern educational landscape (Khanthavit & Khanthavit 2023:215). An excessive reliance on such tools can contribute to heightened academic anxiety, impacting students’ concentration, motivation, and their overall academic performance. In line with the thoughts of Khanthavit and Khanthavit (2023:213), the study underscores the need for a nuanced understanding of the intersection between technological advancements and student wellbeing.

Moreover, the evolution of academic integrity concerns dovetails with the introduction of LLMs such as ChatGPT. Plagiarism detection software and GPT detectors emerge as crucial instruments in identifying and curbing violations. However, challenges persist, especially in detecting content generated by GPTs. This study, thus, highlights the need for IHEs to adapt their strategies to effectively combat emerging forms of academic dishonesty facilitated by advanced AI (Perkins et al. 2023:3 of 21).

Lastly, the study revealed that strategies for fostering academic integrity in this digital epoch necessitate a multifaceted and strategic approach. The recommendations encompass a spectrum of interventions, including educational initiatives, disciplinary measures, policy clarity, technological solutions, and the active involvement of faculty in policy formulation (Azulay Chertok et al. 2014:1328; Roe & Perkins 2022:10 of 10).

Based on the above, we conclude this section on the note that the landscape of academic integrity and the challenges posed by academic anxiety are undergoing a profound transformation in the age of GPT detectors and technological advancements. This synthesis of scholarly insights underscores the intricate interplay between preserving the sanctity of education, addressing student anxiety, and incorporating advanced tools to successfully navigate this complex terrain.

Conclusion and Recommendations

First, academic integrity, which is foundational to the ethos of education, faces both unprecedented threats and promising solutions. The integration of GPT detectors, exemplified by models like GPTZero and CheckGPT, signifies a paradigm shift in combating plagiarism and academic misconduct. While these AI tools offer a robust defense mechanism, their implementation necessitates a delicate balance between surveillance and safeguarding students’ privacy rights. IHEs are challenged to adapt policies and practices to ensure ethical deployment while fostering an environment conducive to creativity and innovation.

Second, the rise of academic anxiety among students poses a significant concern. The digital era, with its relentless competition and elevated expectations, contributes to heightened stress levels and regrettably, unethical practices. GPT detectors, while instrumental in curbing dishonest behaviours, should not be viewed in isolation. They must be part of a comprehensive approach that addresses the root causes of academic anxiety, promoting a culture of trust, support, and resilience.

Third, the multifaceted recommendations from scholarly articles provide a roadmap for IHEs that aim to navigate these challenges effectively. Beyond the technical aspects of implementing GPT detectors, there is a pressing need for educational interventions that enhance students’ understanding of plagiarism, academic integrity policies, and the responsible use of technology. Faculty, as key stakeholder, plays a pivotal role in not only adopting these tools but also in fostering a supportive and understanding academic environment.

Moreover, the fairness and accuracy of GPT detectors in identifying instances of plagiarism are crucial for the credibility of academic assessments. Continuous refinement of detection algorithms, coupled with regular audits and feedback mechanisms, ensures that these tools serve their purpose without disproportionately penalising students. Institutions must be agile in adapting strategies to the evolving tactics employed by students, addressing not only the symptoms but also the underlying causes of academic dishonesty.

As education embraces technological advancements, the onus lies on faculty and IHEs to empower students to navigate this new landscape responsibly. Faculty training becomes indispensable, not just in the technical aspects of using tools like GPT detectors but also in understanding the broader implications for student wellbeing. The intersection of technology and education should be approached with foresight, emphasising the wholistic development of students and creating an environment that encourages critical thinking and genuine learning experiences.

In essence, the conclusion drawn from this exploration is a call for a wholistic and adaptive approach to academic integrity in the digital age. The integration of GPT detectors and technological tools should be complemented by a commitment to addressing the root causes of academic anxiety. By fostering a culture of trust, transparency, and resilience, IHEs can navigate these challenges successfully, ensuring that the pursuit of knowledge remains a transformative and enriching experience for students.

References

Amponsah, S., Van Wyk, MM., & Kolugu, MK. 2022. Academic experiences of ‘zoom-fatigue’ as a virtual streaming phenomenon during the COVID-19 pandemic. International Journal of Web-Based Learning and Teaching Technologies 17(6):1-16. https://doi.org/10.4018/ IJWLTT.287555

Ayyagari, R., Grover, V., & Purvis, R. 2011. Technostress: Technological antecedents and implications. Management Information Systems Quarterly 35(4):831-858. https://doi.org/10.2307/41409963

Ayton, D., Hillman, C., Hatzikiriakidis, K., Tsindos, T., Sadasivan, S., Maloney, S., Bragge, P., Diug, B., & Illic, D. 2022. Why do students plagiarise? Informing higher education teaching and learning policy and practice. Studies in Higher Education 47(9):1921-1934. https://doi.org/10.1080/03075079.2021.1985103

Azulay Chertok, IR., Barnes, ER., & Gilleland, D. 2014. Academic integrity in the online learning environment for health sciences students. Nurse Education Today 34(10):1324-1329. https://doi. org/10.1016/j.nedt.2013.06.002

Belli, S., Raventós, CL., & Guarda, T. 2020. Plagiarism detection in the classroom: Honesty and trust through the Urkund and Turnitin software. In Rocha, Á., Ferrás, C., Montenegro Marin, CE., & Medina García, VH. (Eds.): Information technology and systems, 660-668. Cham: Springer. https://doi.org/10.1007/978-3-030 40690-5_63

Bin-Nashwan, SA., Sadallah, M., & Bouteraa, M. 2023. Use of ChatGPT in academia: Academic integrity hangs in the balance. Technology in Society 75. 102370. 11 pages. https://doi.org/10.1016/j. techsoc.2023.102370

Bretag, T., Harper, R., Burton, M., Ellis, C., Newton, P., Rozenberg, P., Saddiqui, S., & Van Haeringen, K. 2019. Contract cheating: A survey of Australian university students. Studies in Higher Education 44(11):1837-1856. https://doi.org/10.1080/03075079.2 018.1462788

Brown, LA., White, P., Doan, JB., & De Bruin, N. 2011. Selective attentional processing to fall-relevant stimuli among older adults who fear falling. Experimental Aging Research 37(3):330-345. https://doi.or g/10.1080/0361073X.2011.568833

Bultas, MW., Schmuke, AD., Davis, RL., & Palmer, JL. 2017. Crossing the ‘line:’ College students and academic integrity in nursing. Nurse Education Today 56:57-62. https://doi.org/10.1016/j. nedt.2017.06.012

Canzonetta, J. 2021. Repurposing plagiarism detection services for responsible pedagogical application and (in)formative assessment of source attribution practices. Assessing Writing 50, 100563. https://doi.org/10.1016/j.asw.2021.100563

Caporusso, N. 2023. Generative artificial intelligence and the emergence of creative displacement anxiety: Review. Research Directs in Psychology and Behavior 3(1). 12 pages. https://doi.org/10.53520/ rdpb2023.10795

Chatzopoulou, E., Filieri, R., & Dogruyol, SA. 2020. Instagram and bodyimage: Motivation to conform to the ‘Instabod’ and consequences on young male wellbeing. Journal of Consumer Affairs 54:1270-1297. https://doi.org/10.1111/joca.12329

Cotton, DRE., Cotton, PA., & Shipway, JR. 2023. Chatting and cheating: Ensuring academic integrity in the era of ChatGPT. Innovations in Education and Teaching International 61(2):228-239. https://doi. org/10.1080/14703297.2023.2190148

Das, K., 2019. Role of ICT for better mathematics teaching. Shanlax International Journal of Education 7(4):19-28. https://doi. org/10.34293/education.v7i4.641

Eaton, SE. & Gysbers, V. 2021. Academic integrity in online learning: The role of academic librarians. Canadian Journal of Higher Education 51(2):46-64.

Eum, K. & Rice, KG. 2011. Test anxiety, perfectionism, goal orientation, and academic performance. Anxiety, Stress, & Coping 24(2):167 178. https://doi.org/10.1080/10615806.2010.488723

Gimpel, H., Hall, K., Decker, S., Eymann, T., Lämmermann, L., Mädche, A., Röglinger, M., Ruiner, C., Schoch, M., Schoop, M., Urbach, N., & Vandirk, S. 2023. Unlocking the power of generative AI models and systems such as GPT-4 and ChatGPT for higher education. University of Hohenheim . Available at: http://opus.uni hohenheim.de/volltexte/2023/2146/ . (Accessed on 25 October 2023).

Gottardello, D. & Karabag, SF. 2022. Ideal and actual roles of university professors in academic integrity management: A comparative study. Studies in Higher Education 47(3):526-544. https://doi.org/ 10.1080/03075079.2020.1767051

Hagendorff, T., 2020. The ethics of AI ethics: An evaluation of guidelines. Minds and Machines 30(1):99-120. https://doi.org/10.1007/ s11023-020-09517-8

Halupa, C. & Bolliger, DU. 2013. Faculty perceptions of student self plagiarism: An exploratory multi-university study. Journal of Academic Ethics 11(4):297-310. https://doi.org/10.1007/s10805 013-9195-6

Huo, X. & Siau, KL. 2023. Generative AI and ChatGPT impact on technostress of teachers. PACIS 2023 Proceedings. Available at: https://scholars.cityu.edu.hk/en/publications/generative-ai and-chatgpt-impact-on-technostress-of-teachers(8279ae6d c454-4a81-b645-b0edfffe1577).html . (Accessed on 17 November 2023).

International Center for Academic Integrity. 2021. The fundamental values of academic integrity. 3rd ed. International Center for Academic Integrity . Available at: https://academicintegrity.org/resources/ fundamental-values . (Accessed on 17 November 2023).

Jones, LM. & Lee, MJ. 2023. Ethical considerations in automated plagiarism detection: A critical analysis. Journal of Academic Ethics 2023:1-34.

Khanthavit, A. & Khanthavit, S. 2023. ChatGPT and stress. ABAC Journal 43(3):213-224. https://doi.org/10.59865/abacj.2023.39

Kivimäki, M. 1995. Test anxiety, below-capacity performance, and poor test performance: Intrasubject approach with violin students. Personality and individual differences 18(1):47-55. https://doi. org/10.1016/0191-8869(94)00115-9

Kumar, R. 2020. Academic stress and its effect on the academic performance of students. International Journal of Social Sciences and Humanities 4(2):132-141.

Kumar, SP. 2024. TECHNOSTRESS: A comprehensive literature review on dimensions, impacts, and management strategies. Computers in Human Behavior Reports 16. 100475. 16 pages. https://doi. org/10.1016/j.chbr.2024.100475

Leung, L. & Zhang, RW. 2017. Mapping ICT use at home and telecommuting practices: A perspective from work/family border theory. Telematics and Informatics 34(1):385-396. https://doi. org/10.1016/j.tele.2016.06.001

Liang, W., Yuksekgonul, M., Mao, Y., Wu, E., & Zou, J. 2023. GPT detectors are biased against non-native English writers. arXiv 2304.02819v3. 9 pages. Available at: https://arxiv.org/ pdf/2304.02819.pdf . (Accessed on 17 November 2023).

Liu, Z., Yao, Z., Li, F., & Luo, B. 2023. Check me if you can: Detecting ChatGPT-generated academic writing using CheckGPT. arXiv 2306.05524v1. 23 pages.

McCabe, DL. & Pavela, G. 2004. Ten (updated) principles of academic integrity: How faculty can foster student honesty. Change: The Magazine of Higher Learning 36(3):10-15. https://doi. org/10.1080/00091380409605574

Mostofa, M., Tabassum, M., & Ahmed, SMZ. 2021. Researchers’ awareness about plagiarism and impact of plagiarism detection tools – does awareness effect the actions towards preventing plagiarism? Digital Library Perspectives 37(3):257-274. https://doi.org/10.1108/ DLP-10-2020-0100

Perkins, M. 2023. Academic integrity considerations of AI large language models in the post-pandemic era: ChatGPT and beyond. Journal of University Teaching & Learning Practice 20(2). 24 pages. https:// doi.org/10.53761/1.20.02.07

Perkins, M., Roe, J., Postma, D., McGaughran, J., & Hickerson, D. 2023. Game of tones: Faculty detection of GPT-4 generated content in university assessments. arXiv 2305.18081. 21 pages.

Oravec, JA., 2023. Artificial intelligence implications for academic cheating: Expanding the dimensions of responsible human-AI collaboration with ChatGPT. Journal of Interactive Learning Research 34(2):213-237. https://doi. org/10.70725/304731gmmvhw

Qi, C. 2019. A double-edged sword? Exploring the impact of students’ academic usage of mobile devices on technostress and academic performance. Behaviour & Information Technology 38(12):1337 1354. https://doi.org/10.1080/0144929X.2019.1585476

Ragu-Nathan, TS., Tarafdar, M., Ragu-Nathan, BS., & Qiang, T. 2008. The consequences of technostress for end users in organisations: Conceptual development and empirical validation. Information Systems Research 19(4):417-433. https://doi.org/10.1287/ isre.1070.0165

Roe, J. & Perkins, M. 2022. What are automated paraphrasing tools and how do we address them? A review of a growing threat to academic integrity. International Journal for Educational Integrity 18(1). 10 pages. https://doi.org/10.1007/s40979-022-00109-w

Sanghvi, C. 1995. Efficacy of study skills training in managing study habits and test anxiety of high test anxious students. Journal of the Indian Academy of Applied Psychology 21(1):71-75.

Shakir, M. 2014. Academic anxiety as a correlate of academic achievement. Journal of Education and Practice 5(10):29-36.

Sheard, J., Simon,1 Butler, M., Falkner, K., Morgan, M., & Weerasinghe, A. 2017. Strategies for maintaining academic integrity in first-year computing courses. Proceedings of the 2017 ACM Conference on Innovation and Technology in Computer Science Education, 244 249. https://doi.org/10.1145/3059009.3059064

No initials are provided for this author.

Singh, P., Bala, H., Dey, BL., & Filieri, R., 2022. Enforced remote working: The impact of digital platform-induced stress and remote working experience on technology exhaustion and subjective wellbeing. Journal of Business Research 151:269-286. https://doi. org/10.1016/j.jbusres.2022.07.002

Smith, AN. & Smith, BN. 2022. The role of artificial intelligence in maintaining academic integrity. Computers & Education 176. 104298.

Sotiriadou, P., Logan, D., Daly, A., & Guest, R. 2020. The role of authentic assessment to preserve academic integrity and promote skill development and employability. Studies in Higher Education 45(11):2132-2148. https://doi.org/10.1080/03075079.2019.15820 15

Steele, A. 2023. Free speech platforms and the impact of the US insurrection: Misinformation in memes. Master’s thesis, University of Bergen, Bergen. Available at: https://hdl.handle. net/11250/3072107 . (Accessed on 25 November 2023).

Stoesz, BM. & Los, R. 2019. Evaluation of a tutorial designed to promote academic integrity. Canadian Perspectives on Academic Integrity 2(1):3-26. https://doi.org/10.55016/ojs/cpai.v2i1.61826

Turner, SP. & Beemsterboer, PL. 2003. Enhancing academic integrity: Formulating effective honor codes. Journal of Dental Education 67(10):1122-1129. https://doi.org/10.1002 /j.0022-0337.2003.67.10.tb03705.x

Wach, K., Duong, CD., Ejdys, J., Kazlauskaitė, R., Korzynski, P., Mazurek, G., Paliszkiewicz, J., & Ziemba, E. 2023. The dark side of generative artificial intelligence: A critical analysis of controversies and risks of ChatGPT. Entrepreneurial Business and Economics Review 11(2):7-30. https://doi.org/10.15678/ EBER.2023.110201

Weil, MM. & Rosen, LD. 1997. Technostress: Coping with technology @work @home @play. Hoboken: John Wiley.

Whitley, BE. & Keith-Spiegel, P. 2001. Academic integrity as an institutional issue. Ethics & Behavior 11(3):325-342. https://doi. org/10.1207/S15327019EB1103_9

Introduction

AI (artificial intelligence) refers to systems that display intelligent behaviour by analysing their environment and taking actions

– with some degree of autonomy – to achieve specific goals. AI-based systems can be purely software-based, acting in the virtual world (e.g., voice assistants, image analysis software, search engines, speech, and face recognition systems) or AI can be embedded in hardware devices (e.g., advanced robots, autonomous cars, drones, or IoT [internet of things] applications). This definition was proposed by HLEG (the High-Level Expert Group on Artificial Intelligence) in 2019. Samoili, Cobo, Gómez, De Prato, Martínez-Plumed, & Delipetrev (2020:4) suggest that there are two broad categories that can be identified from this definition, namely the ability of machines to reason and make decisions and learn.

There is a transformative force in the era of rapid

advancements in technology that is reforming various sectors, communities, political and educational landscapes, and so on. Nonetheless, this force is transforming the very foundation of human knowledge, and arguably human intelligence. Albert Einstein once stated that ‘the true sign of intelligence is not

knowledge but imagination’ (Ganiev & Tashev 2021:3570).

Einstein went further to argue that imagining is a far more

important skill than knowledge itself. This view implies

that imagination is more important than knowledge thus emphasising the need for imagination alongside knowledge in

fostering creativity. One would posit that AI’s transformative

agenda lies in changing the status quo on knowledge creation.

Jean-Jacques Rousseau’s views on contemporary education and

the role of digital technologies as captured by Jergus (2018:395

396) and Carvalho and Yeoman (2021:28) respectively emphasise

the significance of student’s natural development in the learning

journey and situate learning activity as an emergent process. With this evolution, technological advancement and digital revolution, spearheaded by AI, undoubtedly the longstanding and uncontested dominant status of IHEs (institutions of higher education), in particular universities, in the creation and dissemination of knowledge has now been challenged.

Therefore, drawing from the views of both Einstein and

Rousseau, it is undeniable that AI has reimagined the principle of the knowledge creation process, learning activity, as well as

pedagogical and andragogical strategies. This assertion aligns

with a point made by Eaton (2023:4 of 10) that with AI we have witnessed the era of post-plagiarism, where human creativity is enhanced rather than threatened.

The role of AI has been a subject of discussion both in the mainstream media and academic discourse. The developments in AI are fast changing the educational landscape as these advancements are intimately associated with improvements in computer capabilities, thus resulting in super-intelligent machines. This intelligence is actually the cornerstone of AI as its definition anchors ML (machine learning), whereby computers impersonate humans. Likewise, Trott, Jones, Chang, Michaelov, and Bergen (2023:2 of 21) expound on the human impressionist abilities LLMs (large language models) by asking if LLMs know what humans know. In that piece of work, the authors cite activities such as the ability to learn, adapt, synthesise, and self-correct to complete complex tasks. This happens through LPMs (language processing models). Without a doubt, advancements bring about a lot of opportunities, so there are challenges and associated risks (Neha 2020:308; Popenici & Kerr 2017:1 of 13).

AI’s influence in education builds on and extends what the 4IR (fourth industrial revolution) has done as Oliver (2022b:27) affirms that 4IR alongside IoT, big data, the IIoT (industrial internet of things), and cloud computing cyber-physical systems, to mention a few, has taken the world by storm. Oliver (2022a:53), reflecting on Society 5.0 and Education 5.0, recommends that it is the responsibility of IHEs to lead in innovation incubators and be the cornerstone of ‘sustainable and positive transformation and change.’ This implies, argues Oliver (2022b:27), that IHEs should be instrumental in this era of development given the influence they have in shaping the lives of the youths, their beliefs, skills, future livelihoods, and expectations.

With the advent of ChatGPT (chat generative pre-trained transformer), there have been many other competing and complementary AI tools and systems focusing on computer-generated text, such as Claude, Google Bard, and DALE, to mention a few. These tools fall within the field of Gen-AI (generative artificial intelligence) and can also be used as AWAs (automated writing assistants), AWTs (automated writing tools), and APTs (automated paraphrasing tools) (Roe & Perkins 2022). Other groups of effective AI tools useful for the field of education fall within the realm of IESs (intelligent educational systems) including ITSs (intelligent tutoring systems), educational robotics, and automated grading tools for multimedia systems.

The emergence and launch of ChatGPT in late 2022 by OpenAI has catalysed a rapid rise in open AI technologies and has effectively created an almost instant technological panic about the impact of AI on the HE (higher education) space. There has been worries around technological advancements, while some of the fears centred around diminishing our ability to think (Eaton 2023:2 of 10). Drawing on the developments, (Eaton 2023:2 of

10) adds that the results of technological advancements and by extension AI, are evident in how AI has quickly redefined teaching, learning, and assessment practices.

From a research perspective, advancements in AI present an opportunity for a transformative wave that is reshaping the landscape of academic inquiry across diverse disciplines. Ongoing debates surround the ethical use of AI, with protagonists citing the associated benefits such as enhanced adaptive teaching and assessment, personalised learning experiences, and improved virtual tutoring and support. Among these divergent debates, calls for transformative changes, the need to safeguard the integrity of assessments, and to strike a balance from an ethical viewpoint forms the basis of this chapter.

This chapter adopts a risk-oriented mindset and ethical lens to comprehend how advancements in AI intersect with academic integrity. By dissecting the AI decussation with academic integrity within a HE landscape, the chapter embarks on a journey through the evolving landscape of AI and its profound implications for academic integrity. First, the popularity and development cycles of AI are explored, digging into the hype surrounding Gen-AI and also projecting its future applications. Second, the transformative role of AI, particularly in the health sector is explored. The third section navigates the ongoing debate between AI and human capabilities.

The fourth section closely examines the decussation of AI and academic integrity from the global perspectives of academic dishonesty. This is followed by an exploration of challenges commanded by the rise of a repertoire of LLMs, with special attention to AI authoring tools. In this section, the costs and benefits of AI’s authoring tools are discussed. The positive impacts of the tools such as supporting teaching and learning to negative costs like students’ dependency on AI and plagiarism dangers as well as communication threats arising from reduced human interaction are detailed in the discussion.

The chapter then discourses the proposed strategies to strike a balance between AI benefits and how to maintain academic integrity. As we peer into the future, the second last section reflects on future evolving trends in the AI space in education with commitment challenges. This reflection offers valuable insights for educators, policymakers, and researchers who are constantly navigating the delicate cross paths between technological advancement, specifically AI and academic integrity. The last section provides concluding remarks on the chapter.

AI Popularity and Seasons of Development

AI paradigm shifts can be examined from three distinct but

intertwined phases of development as described by Delipetrev,

Tsinaraki, and Kostic (2020) in their 2020 AI-Watch report for the EU (European Union) Commission’s JRC (Joint Research Centre) examining the evolution of AI. Delipetrev et al. (2020:3) regard the development of AI and its popularity as a seasonal change phenomenon emulating economic cycles as illustrated

in Figure 2.1. These phases are referred to as AI’s seasons of

development and are characterised by sequential stages of upswings (springs for growth) and downswings (winters for

decline) that happen over time. The latter stages are largely due

to cuts in funding opportunities or budgets for development.

The first phase is called AI foundations and occurred between 1950 and 1970. The instrumental pioneers during this primitive era include Alan Turing and John McCarthy. Turing published

what would arguably and undoubtedly be his breakthrough

seminal paper in the field of AI, that debated machines’ ability to think (Turing 1950). The debate continued and was propelled by deliberations at Dartmouth’s conference that took place six years later after Turing’s paper on computing machinery and intelligence. At the conference, McCarthy presented his definition of AI, which he coined in 1955 as ‘the science and engineering of making intelligent machines’ (Delipetrev et al. 2020:7).

Since that conference, excitement about machines, specifically computers’ ability to assist with human challenges, mathematical and science equations, and continued excellence in language translation abilities surged. Furthermore, a myriad of AI initiatives and developments were witnessed – from computer programming, the discovery of perceptron, the realisation of the ELIZA NLP (natural language processing) model to Shakey the Robot (Delipetrev et al. 2020:7). Likewise, Roberts (2016:14) affirms these developments by reflecting that ‘despite the Dartmouth conference’s failure to produce an intelligent machine,’ the aftermath of the conference was nothing but a success. Narrow programmes like the Logic Theorists1 were born, overwhelmingly buying into McCarthy’s optimism and dream, as the government was optimistic about smart machines’ potential to ‘analyze masses of geological data for oil and coal exploration’ (Roberts 2016:14).

The symbolic AI phase ensued between 1970 and 1990. In this era, the underlying discovery was the advent of expert and knowledge systems. The nervous part of these systems was the integration of human beings’ expert knowledge through computer programming. The expert systems were able to perform blood disease diagnostics and prescriptions (Shortliffe & Davis 1975:11), the ID3 (iterative dichotomiser 3) which was able to generate decision trees (Quinlan 1986:82), and the ability for humans to have dialogue with computers (Winograd 1971:75).

The era between 1990 and 2020 indicates that the development of AI took an upward trajectory in an exponential form. The advent of ML and DL (deep learning) has characterised this era, and without a doubt, more is yet to come. The use of convolutional neural networks that read handwritten digits

Logic Theorists is a computer programme written in 1956 by Allen

Newell and two colleagues, Simon and Shaw, engineered to perform

automated reasoning, and has been described as the first artificial

intelligence programme.

(LeCun, Boser, Denker, Henderson, Howard, Hubbard, & Jackel 1989:400), the rise of GANs (generative adversarial networks), and transformers in language modelling for the translation and answering of questions happened in this era. The other instance of AI capabilities surpassing humans includes the 1997 era when IBM (International Business Machines) supercomputer Deep Blue defeated Gary Kasparov in a chess game. No one would have predicted that to happen to the world-renowned world chess master. There is a high degree of uncertainty associated with the future directions of AI and its advancement; however, its maximum potential can only be realised if there is trust in how it is deployed, developed, and used (Thiebes, Lins, & Sunyaev 2021:447).

Notwithstanding all these developments in AI, and it being an important aspect of modern-day life, there are probabilities on the cards where winter seasons in AI development may follow. The same goes for seasons of greater AI spring. This, as Delipetrev et al. (2020:3) accentuate, signifies an uncertainty of the future of AI. Similarly, Schuchmann in his 2019 paper on prospects of an approaching AI winter earlier predicted that given critiques around AI, especially on DL approaches and variations in funding models through R&D (research and development). In fact, on funding R&D, the key role players as far as allocating substantial amounts of funds to AI are concerned, are the US (United States [of America]), China, European Commission through EU member states, India, South Korea, Canada, Japan, Israel, Russia, and Singapore (Samoili et al. 2020). Between 2009 and 2018, there were 34,000 role players globally with the US and China amassing this large concentration.

AI’s Hype Cycles: Focus on Generative AI and Future Expectations

In the history of human evolution, AI is arguably the most intricate and beyond-belief invention where human creativity has been tested and will arguably continue to be tested. Epstein and Hertzmann (2023:1110) affirm this view by stating that new classes of AI tools, especially Gen-AI, have the capability to likely and fundamentally turn around creativity and the idea generation

– from conception to deployment.

AI’s presumed ‘godfather,’ Geoffrey Hinton, a renowned computer scientist who is credited with having built influential and crucial AI tools has shared his crusade on the evolution of AI and the possibility of outsmarting the intelligence of humankind. This was reflected by Joshua Rothman is his article, Why the godfather of AI fears what he’s built, after the intimate interview with Hinton (Rothman 2023). In his reflections, Rothman conveys, what worried Hinton is how humans would prevent AI machines from taking control. His reflection is that no one knows the answer. This evolution and subsequent successes have amplified the debate on machine’s capability to outclass human beings.

Nonetheless, of the debates and successes of machines’ capabilities, the AI field has experienced periods of hype cycles. The AI hype cycle 2023 sheds light on the path and evolution of innovations within AI space, and how they have influenced various sectors (Gartner 2023). Figure 2.2 depicts five phases of the hype cycle characterised by technology triggers, also known as innovation triggers, followed by a peak of an inflated expectation phase.

The third phase relates to the trough of disillusionment, where the ‘hype’ or interest in the innovation starts to wane due to experimental and implementation failures. It is at this stage where the technology fails to meet the expectations. The fourth phase is characterised by a slope of enlightenment where potential benefits of technology start to make sense. The last stage is referred to as a plateau of productivity, where the mainstream adoption of technology takes off and all previous efforts including R&D and marketing start to pay off as Chaffey and Ellis-Chadwick (2019:74) attribute this success to serving the niche market.

Aligned with hype cycles, AI tools are expected to reach a plateau of productivity in four periodic phases. The Grand View Research (2023) report projects the global market size of AI to plateau at almost $2 trillion, a surge of 38.1%. In phase one, tools will reach a plateau in a period of less than two years. This period is followed by the next plateau phase, which is two to five years. In the third phase, a plateau is reached between five and 10 years, while the fourth phase constitutes a period of a decade or more. AI engineering, prompt engineering, AI simulation, and data-centric AI are selected AI fields whose productivity plateaus would mature within two years. Before the impact of AI on academic integrity is explored, Figure 2.3 illustrates how AI is applied and has benefited 11 major industries and sectors, and as Takyar (2022) asserts, AI continues to amass huge potential for the future expansion of these industries.

Some of these breakthroughs have influenced and propelled the discovery of many AI tools in various other sectors, for example, various fields, like the food technology, real estate, health sector, hospitality, travel, and manufacturing (Takyar 2022). The next section selectively explores how AI has transformed the health sector. The choice of this sector is that the results are almost visible as the majority of the people interacts with the sector as patients to consult with their healthcare professionals. This discussion is then followed by a reflection on the big debate between AI and human capability.

Application of AI in the Health Sector

AI has cemented its place as a significant driver for economic development and has been dubbed an ‘area of strategic importance’ given its broader impact on society (Delipetrev et al. 2020:1). The growing influence of AI-driven capabilities in various sectors can be forgiven for making someone believe that they are in a world where they are watching sci-fiction. In this ‘sphere,’ one is experiencing AI algorithms’ superpower and astonishing capabilities. These amazing capabilities include the ability of AI algorithms to compose a masterpiece, which by far outclasses the best classical symphonies of all time. If one thinks of the age of the universe (Planck 2018) and the power of the universe, where Coutinho (2022) refers to the latter as ‘a wonder ahead of medicine,’ which has caught the attention of scientists and researchers and how AI has transcended into this space, the results are astonishing. To date, scientists are still trying to determine the age of the universe as Russ and his colleagues in 1997 concluded that the impact of the inhomogeneities on the global expansion factor was found to be miniature (Russ, Soffel, Kasai, & Börner 1997:2046). They went further to reveal that to use Hubble constantly in the usual way to determine the age of the universe produced an error that was negligible.

According to Manne and Kantheti (2021:78-80), some of the benefits of AI for the health industry include aiding in understanding the needs of patients, cost reduction, and most importantly, improved efficiency. At the heartbeat of this advancement is the availability and processing of healthcare data, which are often voluminous. PWC (2017) cautions that although it is undeniable that AI is gradually becoming more and more sophisticated in performing human activities, the sustainability and accessibility of quality healthcare need to be rethought, given the increasing costs associated with the new discoveries.

In diagnosis and treatment design, the application of and advancement in AI research have been growing and improving. In their 2008 paper, Eren, Subasi, and Coskun (2008) showcase the power of AI in telemedicine through the efficient use of mobile devices and reveal how medical decision support systems can be effective in the provision of quality healthcare services, especially in developing countries that are usually under-resourced and limited by advancement in technologies. Some of the interesting applications of AI in the sector include the deployment of mobile device apps to effectively diagnose pregnant women (Carter, Sandall, Shennan, & Tribe 2019:2) and monitoring nutritional triage for oncology patients (De Bruin, Schuh, Seeling, Luger, Gall, Hütterer, Kornek, Ludvik, Hoppichler, & Schindler 2018:34-35).

The increasing use of AI and robotics in redefining the health sector through examination, diagnosis, and treatment prescription has been witnessed. Exploration of AI chatbots to assist in understanding depression and anxiety in patients (Manne & Kantheti 2021:80) and detecting psychiatric diseases (Erguzel & Ozekes 2014:52) are a few of the AI technologies being tested. Eren et al. (2008:32) suggest that AI techniques like computerised decision support system e.g., ANNs (artificial neural networks), decision trees, and support vector machines are also being used through mobile applications to diagnose and treat problematic diseases like cancer.

In the last decade, the focus was more on a differentiation strategy through explosive innovation in medical products such as newly developed pieces of equipment and consumables. Historical and evidence-based healthcare was influenced by an enormous amount of health data at the disposal of scientists, health practitioners, and regulatory bodies (Frost & Sullivan 2016). These data are in the form of the overwhelming volume of health records, research findings, and diagnosis data that are generated daily. For example, there were approximately 11,000 articles published annually in the field of skincare. One could imagine how this data have assisted dermatologists to more accurately diagnose skin care problems for millions of patients across the globe. In terms of data, the volume of healthcare data was reported to have reached 4 ZBs (zettabytes), considering that one ZB equates a trillion gigabytes. It was projected that these data would exponentially grow by 10 times (PWC 2017:3).

AI versus Humans: The Big Debate!

The field of AI is dynamic and evolving at a fast pace. With its dynamism comes many transformative opportunities. The same can be stated for threats. The latter attests to the need to closely monitor AI through regulatory frameworks, policies, and guidelines. The ample potential impact of AI on humanity is yet to be fully explored. Ironically, the ‘intelligence’ of AI has been debated, and the debate is still ongoing to have the capacity to supersede human intelligence in the near future. In almost affirming this claim, Arrieta, Díaz-Rodríguez, Del Ser, Bennetot, Tabik, Barbado, García, Gil-López, Molina, Benjamins, and Chatila (2020:82) explain that with the manner and speed with which AI-powered systems are sophisticated, the advancement has led to a situation in which these systems can act or operate independently of human intervention, especially in their design and deployment.

For example, in the health sector, Manne and Kantheti (2021:78) are pondering whether the availability and use of AI tools in the future will replace physicians. The authors quickly respond by stating that practically, it is impossible. At most, however, these tools will enable physicians to achieve improved and accurate results. The vantage point lies in its ability to process enormous amounts of information a billion times faster, thus enabling it to decide much quicker than human beings can. Lu (2023), in an article headlined How smart is ChatGPT? reveals riveting results on how it is able to compete with humans in examinations. These results are worth reflecting on. In the latest release, ChatGPT 4 is reportedly capable of matching the human performance level in most of professional and academic assessments. Despite these breakthroughs, it is still struggling in subjects like English language and literature (Lundgren 2024; Nasrullah & Wahyu 2024). ChatGPT 4’s performance in language tasks is still exhibiting inconsistencies, thus suggesting its insufficient grasp of language and literature intricacies. The subjects in which it is demonstrating high grading and comprehension accuracy and it is performing well are in the STEM (science, technology, engineering, and mathematics) field with subjects like physics, statistics, mathematics, and biology (Lundgren 2024; Ragolane, Patel, & Salikram 2024). This is particularly true when questions are objective i.e., multiple-choice or sort-answer.

The often mystic and evolving incredible capabilities of AI and its algorithms fuel the ‘big debate’ on intelligence about AI versus human intelligence, as discussed in this section. The question of whether AI can surpass human intelligence is explored with questions like the following:

The questions lean on the protagonists who are on the side of those arguing that AI will replace humans in their jobs. The debate is not new as advancements in technology and the prevalence of automated machines, robots, and production processes, especially in the automotive industries and huge supply chains have replaced human beings. In an attempt to respond to the first and last questions, I argue that the potential of technology, especially AI and data science, to benefit humanity is substantial, yet it is not without challenges. Various studies (e.g., Beduschi 2024; Thiebes et al. 2021) highlight how these technologies can contribute positively to societal goals, but they also caution against the risks of inequality and ethical concerns. Hence, there is a quest in some parts of the literature geared towards promoting responsible and sustainable AI in protection of human rights against bias and discrimination in algorithms, whether intentional or otherwise (Lamchek 2023; Yadav, Tudela, & Marco-Lajara 2024). With reference to the second question, sections 2 and 3 focused on the capabilities of AI and its impact in various sectors such as health, education and labour market among others.

It is undeniable that AI has transcended and has been transformative in enriching the human experience. Contemporary advancement in AI and its applications in various industries and sectors of the economy is unprecedented. In this section, the impact of AI in the health sector is highlighted. The healthcare sector experienced outstanding AI advancements that are on the brink of revolutionising patient care and outcomes. Advancing medical diagnostics through image recognition and predictive analytics, enabling early disease detection, and personalised treatment plans are some of the developments AI has advanced in the industry. The influence of AI in the health sector was evidenced in 2012 when Geoffrey developed a neural network at the University of Toronto with two of his students. The system they have built, was able to mimic humans’ behaviour and function just like the human brain (Apperly & Butterfill 2009:953) where one could feed it with a large amount of data, and through ML it could learn how to identify objects with almost near perfect efficacy.

The working paper of the ILO (International Labour Organization) on the global analysis of jobs that are exposed to AI, specifically Gen-AI, projects that the effects vary across countries based on their income level (Gmyrek, Berg, & Bescond 2023). For example, countries with low-income levels have 0.4% of their total labour force exposed compared to 5.5% of countries in the upper bound. Administrative jobs that are clerical in nature are the most exposed at 24%. Meanwhile, other occupational groups vary between one and four percent (Gmyrek Berg, & Bescond 2023:110). Along the same lines, the OECD (Organisation for Economic Cooperation and Development) (OECD 2023:42) suggests that AI is changing the jobs landscape and estimates that 42% and 47% of automated jobs (AI adoption) in finance and manufacturing respectively, were motivated by reduction in staff costs.

The quote by Kevin Kelly, co-founder of Wired magazine in his book, The inevitable shares rather interesting views on the future of jobs in the world of AI: ‘Your job won’t be taken by AI, but it will be taken by someone who uses AI’ (Kelly 2016:2). At the core of Kelly’s view is that AI is not here to replace people in their jobs, however, there is a need for adaptability and an urge for self-fulfilment and motivation in one’s career development. Likewise, in an interview with CNBC (2023), Elon Musk, the CEO of X, a leading social media platform previously known as Twitter, has advised people, youths in particular, to embark on a career journey that keeps them captivated and always satisfied. This background in a way opines that there are varied views on the pros and cons of AI evolution. Now that the impact of AI has been discussed alongside the great debate on AI versus humans, in the next section the decussation of AI and academic integrity is presented.

AI on Academic Integrity Decussation: Plausibility for Evolution of Academic Dishonesty and Academic Misconduct

There is no universal definition of academic dishonesty, just as there is none for plagiarism. Both the concepts of academic dishonesty and academic misconduct are often used interchangeably, referring to the acts of violating academic integrity (Eaton 2017:272). Alongside plagiarism, the concepts have dominated ethical discourse in IHEs (Hua 2023:160). Eaton (2017:272) continues to suggest that academic dishonesty and misconduct are regarded as forms of an academic integrity antithesis. This affirms what the ICAI (International Centre for Academic Integrity) on the fundamental values of academic integrity stated, viewing both academic dishonesty and misconduct as compromising essential ethical principles and values of trust, honesty, responsibility, fairness, and respect (ICAI 2021:4-10). Tracey Bretag, the Australian scholar, who is popular for, and an avid advocate of academic integrity (cf. Marais 2022:57), borrows from the definition of APFEI (assessment of process for efficiency improvement) of academic integrity as follows: ‘A commitment to key values of honesty, trust, fairness, equity, respect and responsibility, and the translation of these values into action’ (Bretag 2016:28).

The popularity of the term ‘academic integrity’ is credited to Donald McCabe, Rutgers Business School’s Professor of Management and global Business, who is dubbed to be its father and to some known as ‘Dr. Ethics’ (Rutgers Business School 2014). In his summer reflection on liberal education, McCabe (2005:26) writes that ‘it takes a village: Academic dishonesty and educational opportunity [and argues that] if we truly believe in our role as educators, we would do better to view most instances of cheating as educational opportunities.’ There are other prominent scholars who have made and some still are making an immense contribution to the field of academic integrity.2 Enacting behavioural principles and values is the backbone of academic integrity, while the aim is to avoid acts of plagiarism and contract cheating, as Clarke and Lancaster (2006:19) as well as Eaton (2018) put it. In affirmation, Kirk (1996:16) continues to argue that maintaining academic integrity upholds high standards of honesty and rigour in academic research and publishing.

2 Examples are Tracey Bretag and Cath Ellis, both from Australia,

Tricia Bertram Gallant from the US, Sarah Elaine Eaton from Canada,

Thomas Lancaster from the UK, and Tomáš Foltýnek from the Czech

Republic.

On the backdrop of proliferation in AI, the growing importance of protecting and maintaining the practice of a moral or principled code of ethical behaviour, which is the core value of academic integrity is concomitantly taking centre stage (Marais 2022:57). Since the COVID-19 (Coronavirus disease of 2019) pandemic, IHEs across the globe had to, willingly and unwillingly, transition to online assessments (Perkins 2023:1 of 24). Inevitably, this transition was accompanied by concerns on the rising prevalence of academic dishonesty. To comprehend the complexity of academic dishonesty, it is important to note the origins of assessments, which date back as far as the primitive era of the first tests3 in the 18th century (Lang 2013:14). Time-honoured methods such as plagiarism, unauthorised collaboration, and exam cheating remain prevalent challenges. Students have, historically, sought to gain an unfair advantage through these means, raising concerns for educators and institutions globally. The use of crib sheets, copying from peers, and recycling assignments have been persistent issues, emphasising the need for ongoing vigilance in upholding academic integrity (Armstead 2001:10; Stevens 2013:17).

It has been argued that academic integrity tends to place the responsibility of good ethical conduct on the individual, which, according to Fishman (2015:13-16) in her book chapter on Academic integrity as education concept: Concern, and movement in US institutions of higher learning assumes that approach to be problematic. Other scholars point to the Chinese civil service over a millennium when the Chinese emperor in early 2200 BCE administered an elementary form of testing for his officials once every three years as a measure of assessing their fit-for-office (Bushway & Nash 1977:4). In the context of the US, Gregory (2004:59) documents the late 1800s to be the era during which tests were popularised. James McKeen Cattell was credited to have imported brass testing instruments to the US; hence he

According to Fishman (2015:7-8), the historical evolution of academic integrity shaped American approaches to academic integrity and can be traced back to the early conceptions, which were based on British

higher education models. These models explicitly addressed ethical and moral lessons through explicit, religious beliefs and mandatory

practices.

earned an accolade of being referred to as ‘undisputed dean of American psychology’ (Gregory 2004:60-61). This long and distinct period of history was preceded by the Francis Galton era (1822-1911). Galton who was reported to have been obsessed with measurements, devised the first battery of tests within the field of psychology in Great Britain. Akin to the introduction of any other system, administering the assessments was not without challenges, and this led to testing environments needing refinements and modifications over centuries. It was until this time that written exams were enacted in the Han dynasty. Some of the challenges relates to unnecessary demanding assessment practices and invalidated selections tests (Gregory 2004:57). Other challenges include cheating and bribing of examiners, which were popular despite these acts carrying the death penalty as a form of deterrent for both examiner and test taker should it be proven that they are guilty of misconduct (Bushway & Nash 1977:623; Gregory 2004:57-59; Zhang 2017:153). The next section takes a closer view on the contemporary challenges in academic integrity from the context of pre- and post-COVID-19 pandemic.

Global Perspectives on Academic Integrity

This section explores the contemporary issues surrounding the management of academic integrity, looking at both the conventional forms and evolving forms of academic dishonesty. Additionally, we examine this through contrasting the global and local perspectives on academic integrity. The focus of this exercise is on selected developed countries contrasted with emerging nations like Brazil, India, and South Africa. The aim of this comparative analysis is to shed light on the complex consequences of academic integrity issues in various learning environments and educational landscapes.

The challenges in maintaining are on the one hand precipitated by perspectives on the development of academic integrity measures, while on the other there is the role played by technology. On the former, Fishman (2015:8) attributes and relates challenges to the narrow view of academic integrity. Authors go further to suggest that there needs to be a paradigm shift away from students as possible threats on academic integrity, towards focusing on educators and researchers, practices, and entrenched institutional ethical cultures. The shortcoming of this one-dimensional perspective rests on the responsibility of managing academic integrity on individual choices, hence the inefficiencies. Therefore, a multi-dimensional perspective of responsibility of managing academic integrity, a principle which this chapter is underpinned on, remains the effective and responsible way of managing academic integrity.

Developed nations like the US, UK (United Kingdom), Italy, and Australia often face challenges associated with technology-driven academic misconduct. For the longest time, academic integrity has been the anchor behind fair, valid, and reliable education system. The assurance of fair and honest representation of one’s work in assessments underpins the basis for principles of academic integrity. Sadler (1989:119), in the theory of formative assessment, suggests that there are multiple criteria applied in judging the quality of students’ assessments. It is the ‘quality’ in this assessment that is a beacon of hope when it comes to protecting academic integrity. However, owing to evolutions in the educational landscape, the course or journey to manage and maintain academic integrity, is by extension raddled with allied challenges.

Current Challenges in Academic Integrity

Using AI tools does not automatically constitute academic dishonesty. Eaton (2023:3 of 10) uses the six tenets of post-plagiarism by reflecting writing in the age of AI and argues that attempts to detect where the human ends and where the AI begins in this era is pointless. Examining academic integrity on a global scale reveals a spectrum of practices and attitudes. The proliferation of essay mills and the use of advanced cheating technologies pose threats to the integrity of assessment processes. On the other hand, developing countries such as Brazil, India, and South Africa grapple with issues rooted in resource constraints and varying levels of academic support. The implications for educational institutions in these diverse contexts are profound, necessitating tailored strategies to address the unique challenges each faces.

Academic integrity literature provides valuable insights into the implications and challenges in academic integrity, both in developed and developing nations. In both the US, Canada, and the UK, studies highlight the prevalence of contract cheating and the need for rigorous academic policies (Eaton 2018; Eaton, Rothschuh, Fernández Conde, Guglielmin, Otoo, Wilson, & Burns 2018:2; Eaton 2017:271; Lancaster 2020:1-2 of 14; Stone 2023:357). Perkins (2023:2 of 24) suggests that in Italy and Australia the focus has shifted towards the impact of cultural attitudes on academic honesty and considerations of academic integrity in the era of AI LLM models (Ellis, Van Haeringen, Harper, Bretag, Zucker, McBride, Rozenberg, Newton, & Saddiqui 2020:454; Bretag 2016:8-10). Conversely, in India and South Africa, the reviewed research underscores the role of socioeconomic factors in influencing academic misconduct (Marais 2022; Mwamwenda 2006; Vandana & Nagaveni 2019).

Key implications gleaned from the literature include the urgent need for standardised global frameworks that acknowledge the contextual differences between developed and developing nations. Institutions must not only adopt proactive measures to deter academic misconduct but also foster a culture of academic honesty through educational initiatives. As we navigate the post-COVID educational landscape, understanding these implications becomes paramount in crafting effective strategies to safeguard academic integrity worldwide.

The Costs of the Rise in AI’s Large Language Models Authoring Tools on Academic Integrity

The advent of technology has ushered in a new era of challenges for academic integrity. While traditional forms of cheating persist, there is a notable evolution in the methods employed. According to Epstein and Hertzmann (2023:1110), the capability and influence of Gen-AI tools lead to reimagining the creativity, as such, posing a need to understand the impact as it cuts across many sectors of the society. The HE landscape, and by extension, academic integrity is no exception. For example, one of the concerns raised against LLMs is their ability to generate text that may be difficult for academics to detect that it is machine-generated (Perkins 2023:8 of 24; Wahle, Ruas, Kirstein, & Gipp 2022:952). Another challenge is the difficulty with which to establish relevant policies if the use of LLMs as a tool for student cognitive offloading is considered. LLMs are transformer-based ML models characterised from language-based activities (Vaswani, Shazeer, Parmar, Uszkoreit, Jones, Gomez, Kaiser, & Polosukhin 2017). In this section, the key question under exploration is, ‘To what extent does the rise in AI authoring tools compromise academic integrity?’

The evolution of AI academic authoring tools has its own ‘the good and the bad’ story. Much as there are benefits – intended or unintended – there are also associated consequences or costs. These costs are propelling the debates and arguments for and against embracing AI. Grappling with the challenges brought by the rise in AI authoring tools, universities and other IHEs around the world have recently developed (and are still developing, whilst others are yet to develop) their policy stances on the usage of AI authoring tools. On the backdrop of the downside of the rise in AI authoring tools, the likes of the AWG (Australian Writers’ Guild) have in its response to this explosion, released a statement where the association vehemently affirms its stringent stance against the unregulated use of AI within Australia’s creative industries (AWG 2023). Similarly, the WGA (Writers Guild of America) has tailed its concerns and issues they have on AI within the US screen industry, and this went as far as the protracted protests. The WGA ultimately won critical contract protections, although this victory was not enough, according to Winter (2023). Despite the critics, at the core of the settlement, it was agreed that AI would not be used to weaken writers’ credits and compensation. In another prominent example of concerns against AI, the technology reporter of the BCC (British Broadcasting Corporation), Shiona McCallum wrote that ‘ChatGPT [was] banned in Italy over privacy concerns’ (McCallum 2023). Unexpectedly and ironically, the headline would coincide with the April fool’s day, and uncertainly some would wonder if this was a gimmick. However, it was found to be a true story. As reported in the article, the data protection authority in Italy was grimly concerned about privacy of the data on the chatbot’s model (McCallum 2023).

To borrow from the economics literature, there is a popular phrase about the hidden or opportunity costs that states: ‘There ain’t no such thing as a free lunch’, to which Safire William (1993) argues that it was popularised by the free market economist Milton Friedman in 1975 (Friedman 1975). This phrase puts meaning and understanding to the existence and impact of costs in any transaction. Brue and McConnell (2014:3) put it succinctly as they write that the phrase is ‘at the core of economics.’ It is undeniable that much as there are benefits brought by AI authoring tools, there are also concomitant costs. The impact of the LLMs on academic integrity is complex and is continuously evolving. This is partly attributed to unremitting development and deployment of new tools. For example, Poola (2023:16) states that with the development of new technologies and advancement in AI, the efficiency and effectiveness of LLMs such as ChatGPT is likely to be impacted. This is more of a paradox situation with positive and negative potential consequences and will be discussed in the next sections.

Positive Benefits of AI Authoring Tools

The rise of AI authoring tools represents a transformative wave that is reshaping the landscape of academic inquiry across diverse disciplines (Currie 2023:723). Notwithstanding the negative consequences of AI authoring tools which are discussed in the next section, the positive impact of AI on academic integrity leans towards supporting and enhancing the individual’s learning experience. The focus of the next two sections is to demonstrate how AI can be positively used to support learning. This is followed by a discussion on how AI systems can collaborate with human beings.

Future Trends: Evolution of AI and Related Challenges

The future of AI and its continuous advancements remain uncertain. This evolution is dynamic in nature and will undoubtedly continue to bring many surprises. In turn, the future of academic integrity stands at a crossroads, marked by the transformative influence of AI and the challenges it introduces. As we propel into an era characterised by reduced human interaction, communication threats, students’ increasing dependency on AI, and ethical concerns stemming from heightened academic dishonesty, it becomes imperative to recalibrate our strategies and fortify the foundations of academic integrity.

The reduction in human interaction, exacerbated by the widespread adoption of digital platforms and automated assessment tools, poses a communication threat that widens the gap between educators and students. The traditional channels for guidance and mentorship may dwindle, necessitating innovative approaches to foster meaningful connections. Educators and institutions must proactively leverage technology to bridge this gap, establishing clear lines of communication, virtual support systems, and avenues for students to seek guidance and feedback, especially on clarifying the difference between plagiarism and poor referencing techniques (Stone 2023:372).

The growing dependency on AI among students raises concerns about diminishing capacities for critical thinking. As AI tools streamline processes and offer ready-made solutions, there is a risk of stifling the development of essential analytical and problem-solving skills. The future demands a recalibration of educational approaches to strike a balance between leveraging AI for efficiency and preserving the space for students to cultivate critical thinking. Encouraging inquiry-based learning, complex problem-solving, and projects that require creative thought will be crucial in nurturing students’ intellectual independence in attempting to overcome transparency issues in the current available LLMs, alternative open source LLMs known as OPTs (open pre-trained transformers) are being developed as they boost the advantage of a text generation ability and are more open to scrutiny (Perkins 2023:7 of 24).

Conclusion

In the face of changing educational environments and advancements in AI, the survival of academic integrity depends on the current and future generation’s capacity to adjust, create, and preserve instead of subverting moral principles. To fully utilise AI’s benefits while preserving the values that guide academic success, a deliberate effort is needed. There is a need to create an educational future that is both technologically advanced and morally sound. This involves tackling issues presented by a worryingly mounting students’ dependency on AI, less human interaction, communication, and ethical gaps. The journey to maintaining academic integrity in the AI age remains a dynamic call that requires ongoing self-examination, adjustment, and a resolute dedication to the values that constitute the core of intellectual endeavours. Eaton (2023) thus argues that we need to be talking about post-plagiarism and reflects on ethics and integrity in the era of AI.

Equally so, the ethical threat posed by increased academic dishonesty in the digital age demands a proactive and multifaceted response. IHEs must not only employ advanced technological tools for plagiarism detection, as they need to up their investment game as far as comprehensive educational programmes that instil a strong sense of academic ethical culture are concerned. Fostering a culture of integrity, emphasising the value of original thought, and equipping staff – students, educators, and management – with the skills to navigate the ethical challenges posed by AI are essential steps for mitigating the possible threat against academic integrity.

References

Almazroa, H. & Alotaibi, W. 2023. Teaching 21st century skills: Understanding the depth and width of the challenges to shape proactive teacher education programmes. Sustainability 15(9). 7365. 25 pages. https://doi.org/10.3390/su15097365

Apperly, IA. & Butterfill, SA. 2009. Do humans have two systems to track beliefs and belief-like states? Psychological Review 116(4):953 970. https://doi.org/10.1037/a0016923

Armstead, PK. 2001. What factors affect cheating in secondary school and why? PhD thesis, Department of Psychology, University of Plymouth, England.

Arrieta, AB., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., García, S., Gil-López, S., Molina, D., Benjamins, R., & Chatila, R. 2020. Explainable artificial intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information fusion 58:82-115. https://doi. org/10.1016/j.inffus.2019.12.012

AWG (Australian Writers’ Guild). 2023. Australian Writers’ Guild rejects unregulated AI in the creative sector. 28 September 2023. Available at: https://www.awg.com.au/News-Details?newsId=8 93&title=Australian-Writers-Guild-rejects-unregulated-AI-in the-creative-sector . (Accessed on 25 December 2023).

Beduschi, A. 2024. Synthetic data protection: Towards a paradigm change in data regulation? Big Data & Society 11(1):20-53. https://doi. org/10.1177/20539517241231277

Bouchard, L. 2022. Prompting explained: How to talk to ChatGPT. 12 December 2022. Available at: https://www.louisbouchard.ai/ prompting-explained/ . (Accessed on 2 January 2024).

Bretag, T. 2016. Handbook of academic integrity. Singapore: Springer. https://doi.org/10.1007/978-981-287-098-8

Brue, S. & McConnell, C. 2014. Essentials of economics. New York: McGraw Hill.

Bushway, A. & Nash, WR. 1977. School cheating behaviour. Review of Educational Research 47(4):623-632. https://doi. org/10.3102/00346543047004623

Carter, J., Sandall, J., Shennan, AH., & Tribe, RM. 2019. Mobile phone apps for clinical decision support in pregnancy: A scoping review. BMC Medical Informatics and Decision Making 19(1):1-13. https://doi. org/10.1186/s12911-019-0954-1

Carvalho, L. & Yeoman, P. 2021. Performativity of materials in learning: The learning-whole in action. Journal of New Approaches in Educational Research 10(1):28-42. https://doi.org/10.7821/ naer.2021.1.627

Centre for AI Safety. 2023. Statement on AI risk: AI experts and public figures express their concern about AI risk. Centre for AI Safety. Available at: https://www.safe.ai/statement-on-ai-risk. (Accessed on 7 January 2024).

Chaffey, D. & Ellis-Chadwick, F. 2019. Digital marketing. Harlow: Pearson.

Chassignol, M., Khoroshavin, A., Klimova, A., & Bilyatdinova, A. 2018. Artificial intelligence trends in education: A narrative overview. Procedia Computing Science 136:16-24. https://doi.org/10.1016/j. procs.2018.08.233

Clarke, R. & Lancaster, T. 2006. Eliminating the successor to plagiarism? Identifying the usage of contract cheating sites. Proceedings of 2nd international plagiarism conference, June 2006, 19-21.

CNBC. 2023. Tesla CEO Elon Musk discusses the implications of A.I. on his children’s future in the workforce. 16 May 2023. Video. Available at: https://www.cnbc.com/video/2023/05/16/tesla-ceo-elon musk-discusses-the-implications-of-a-i-on-the-future-of

the-workforce.html . (Accessed on 3 January 2024).

Coppin, B. 2004. Artificial intelligence illuminated. Boston: Jones & Bartlett.

Coutinho, L. 2022. The power of the universe: How to make it work for you. The Free Press Journal . 9 April 2022. Available at: https:// www.freepressjournal.in/weekend/the-power-of-the-universe how-to-make-it-work-for-you . (Accessed on 9 December 2023).

Currie, GM. 2023. Academic integrity and artificial intelligence: Is ChatGPT hype, hero or heresy? Seminars in Nuclear Medicine 53(5):719-730. https://doi.org/10.1053/j.semnuclmed.2023.04.008

De Bruin, JS., Schuh, C., Seeling, W., Luger, E., Gall, M., Hütterer, E., Kornek, G., Ludvik, B., Hoppichler, F., & Schindler, K. 2018. Assessing the feasibility of a mobile health-supported clinical decision support system for nutritional triage in oncology outpatients using Arden Syntax. Artificial Intelligence in Medicine 92:34-42. https://doi.org/10.1016/j.artmed.2015.10.001

Delipetrev, B., Tsinaraki, C., & Kostic, U. 2020. Historical evolution of artificial intelligence. Luxembourg: Publications Office of the European Union.

Dong, Z. 2023. Research on the impact of artificial intelligence on the development of education. Proceedings of the 2nd International Conference on Interdisciplinary Humanities and Communication Studies.

Eaton, SE. 2017. Comparative analysis of institutional policy definitions of plagiarism: A Pan-Canadian university study. Interchange 48:271-281. https://doi.org/10.1007/s10780-017-9300-7

Eaton, SE. 2018. Contract cheating: A Canadian perspective. BMC. 24 July 2018. Available at: http://blogs.biomedcentral.com/ bmcblog/2018/07/24/contract-cheating-a-canadian perspective/ . (Accessed on 29 November 2023).

Eaton, SE. 2023. Postplagiarism: Transdisciplinary ethics and integrity in the age of artificial intelligence and neurotechnology. International Journal for Educational Integrity 19(1). 23. 10 pages. https://doi.org/10.1007/s40979-023-00144-1

Eaton, SE., Rothschuh, S., Fernández Conde, C., Guglielmin, M., Otoo, BK., Wilson, J., & Burns, I. 2018. Understanding faculty perceptions and approaches to academic integrity in a Canadian school of education. Available at: https://files.eric.ed.gov/fulltext/ ED581141.pdf . (Accessed on 4 January 2024).

Ellis, C., Van Haeringen, K., Harper, R., Bretag, T., Zucker, I., McBride, S., Rozenberg, P., Newton, P., & Saddiqui, S. 2020. Does authentic assessment assure academic integrity? Evidence from contract cheating data. Higher Education Research & Development 39(3):454-469. https://doi.org/10.1080/07294360.2019.1680956

Epstein, Z. & Hertzmann, A. 2023. Investigators of human creativity. In Akten, M., Farid, H., Fjeld, J., Frank, MR., Groh, M., Herman, L., Leach, N., & Mahari, R. (Eds.): Art and the science of generative AI. Science 380(6650):1110-1111. https://doi.org/10.1126/science. adh4451

Eren, A., Subasi, A., & Coskun, O. 2008. A decision support system for telemedicine through the mobile telecommunications platform. Journal of Medical Systems 32(1):31-35. https://doi.org/10.1007/ s10916-007-9104-x

Erguzel, TT. & Ozekes, S. 2014. Artificial intelligence approaches in psychiatric disorders. The Journal of Neurobehavioral Studies 1:52 53. https://doi.org/10.5455/JNBS.1405259279

Ettinger, RCW. 1965. The prospect of immortality. London: Sidgwick and Jackson.

Ettinger, RCW. 2005. Man in superman: The startling potential of human evolution – and how to be part of it. New York: Routledge.

Fajet, W., Bello, M., Leftwich, SA., Mesler, JL., & Shaver, AN. 2005. Pre-service teachers’ perceptions in beginning education classes. Journal of Teaching Education 21(6):717-727. https://doi. org/10.1016/j.tate.2005.05.002

Fan, W., Chen, Y., Li, J., Sun, Y., Feng, J., Hassanin, H., & Sareh, P. 2021. Machine learning applied to the design and inspection of reinforced concrete bridges: Resilient methods and emerging applications. Structures 33:3954-3963. https://doi.org/10.1016/j. istruc.2021.06.110

Fishman, T. 2015. Academic integrity as an educational concept, concern and movement in US institutions of higher learning. In Bretag, T. (Ed.): Handbook of academic integrity, 7-22. Singapore: Springer. https://doi.org/10.1007/978-981-287-098-8_1

Friedman, M. 1975. There’s no such thing as a free lunch. Chicago: Open Court Publishing Company.

Frost, K. & Sullivan, B. 2016. Transforming healthcare through artificial intelligence systems. Artificial Intelligence in Health and Life Sciences. DocPlayer . 19 slides. Available at: https://docplayer. net/36848717-Transforming-healthcare-through-artificial intelligence-systems.html . (Accessed on 30 December 2023).

Fuchs, T. 2021. Beyond the human? A critique of transhumanism. In Fuchs,

T. (Ed.): In defence of the human being: Foundational questions of an embodied anthropology, 49-82. Oxford: Oxford University Press. https://doi.org/10.1093/oso/9780192898197.003.0003

Ganiev, AG. & Tashev, SN. 2021. The role of ‘imagination’ in the process of ‘creative thinking’ developing students’ ‘imagination’ and ‘creative thinking’ skills in teaching physics. Psychology and Education 58(1):3569-3575. https://doi.org/10.17762/pae. v58i1.1309

Gartner, 2023. What’s new in artificial intelligence from the 2023 Gartner hype cycle. 17 August 2023. Available at: https://www.gartner. com/en/articles/what-s-new-in-artificial-intelligence-from the-2023-gartner-hype-cycle . (Accessed on 5 December 2023).

Gmyrek, P., Berg, J., & Bescond, D. 2023. Generative AI and jobs: A global analysis of potential effects on job quantity and quality. ILO Working Paper 96. Geneva: ILO. https://doi.org/10.54394/FHEM8239

Grand View Research. 2023. Digital transformation market to reach $4,617.78 billion by 2030. 8 August 2023. Available at: https:// www.grandviewresearch.com/press-release/global-digital transformation-market# . (Accessed on 7 December 2023).

Gregory, RJ. 2004. Psychological testing: History, principles, and applications. Boston: Allyn & Bacon.

Hellman, M. 2019. Augmented writing technology: A writer’s friend or foe? The Seattle Times . 30 April 2019. Available at: https:// www.seattletimes.com/business/technology/augmented writing-technology-a-writers-friend-or-foe/ . (Accessed on 30 December 2023).

Hua, JH. 2023. Beyond exams: Investigating AI tool impact on student attitudes, ethical awareness, and academic dishonesty in online college assessments. International Journal of Educational Management and Development Studies 4(4):160-185. https://doi. org/10.53378/353030

Huxley, J. 1957. Transhumanism. London: Chatto & Windus. ICAI (International Centre for Academic Integrity). 2021. The fundamental values of academic integrity. 3rd ed. Albany: ICAI.

Jergus, K. 2018. Jean-Jacques Rousseau (1712-1778). In Smeyers, P (Ed.): International handbook of philosophy of education, 395-406. Part 1. Cham: Springer. https://doi.org/10.1007/978-3-319-72761-5_33

Jones, M. & Sheridan, L. 2015. Back translation: An emerging sophisticated cyber strategy to subvert advances in ‘digital age’ plagiarism detection and prevention. Assessment & Evaluation in Higher Education 40(5):712-724. https://doi.org/10.1080/02602938.201 4.950553

Kaartemo, V. & Helkkula, A. 2018. A systematic review of artificial intelligence and robots in value co-creation: Current status and future research avenues. Journal of Creating Value 4(2):211-228. https://doi.org/10.1177/2394964318805625

Kamiya, A. 2023. Transhumanism: History, criticisms & future of humanity. Study.com . 8 March 2023. Available at: https://study. com/academy/lesson/transhumanism-overview-history debates.html . (Accessed on 2 January 2024).

Kelly, K. 2016. The inevitable: Understanding the 12 technological forces that will shape our future. New York: Penguin.

Khan, MA. & Vivek, MA. 2022. Artificial intelligence and big data: The advent of new pedagogy in the adaptive e-learning system in the higher educational institutions of Saudi Arabia. Education Research International 2022. 10 pages. https://doi. org/10.1155/2022/1263555

Kirk, A. 1996. Learning and the marketplace: A philosophical, cross-cultural (and occasionally irreverent) guide for business and academe. Carbondale: Southern Illinois University Press.

Korteling, JH., Van de Boer-Visschedijk, GC., Blankendaal, RA., Boonekamp, RC., & Eikelboom, AR. 2021. Human- versus artificial intelligence. Frontiers in Artificial Intelligence 4. 622364. 13 pages. https://doi.org/10.3389/frai.2021.622364

Kurosu, M. (Ed.): Human-computer interaction: Theory, methods and tools. HCII 2021. Lecture Notes in Computer Science 12762. Cham: Springer. https://doi.org/10.1007/978-3-030-78462-1

Lamchek, JS. 2023. Ensuring data science and its applications benefit humanity: Data monetization and the right to science. Human Rights Law Review 23(3):1-23. https://doi.org/10.1093/hrlr/ ngad018 Lancaster, T. 2020. Commercial contract cheating provision through micro-outsourcing web sites. International Journal for Educational Integrity 16(4). 14 pages. https://doi.org/10.1007/s40979-020 00053-7

Lang, JM. 2013. Cheating lessons: Learning from academic dishonesty. Cambridge: Harvard University Press.

LeCun, Y., Boser, B., Denker, JS., Henderson, D., Howard, RE., Hubbard, W., & Jackel, LD. 1989. Handwritten digit recognition with a back-propagation network. In Touretzky, D. (Ed.): Advances in neural information processing systems, 396-404. Part 5. NIPS Conference. Denver: Morgan Kaufmann.

Leslie, D. 2019. Understanding artificial intelligence ethics and safety: A guide for the responsible design and implementation of AI systems in the public sector . London: The Alan Turing Institute. https://doi. org/10.2139/ssrn.3403301

Lu, M. 2023. How smart is ChatGPT? Visual Capitalism. 26 April 2023. Available at : https://www.visualcapitalist.com/how-smartis-chatgpt/#:~:text=ChatGPT%2C%20a%20language%20 model%20developed,help%20them%20with%20their%20 homework. (Accessed on 12 December 2023).

Lundgren, M. 2024. Large language models in student assessment: Comparing ChatGPT and human graders. arXiv preprint. https:// doi.org/10.2139/ssrn.4874359

Mahabeer, P. & Pirtheepal, T. 2019. Assessment, plagiarism and its effect on academic integrity: Experiences of academics at a university in South Africa. South African Journal of Science 115(11-12):1-8. https://doi.org/10.17159/sajs.2019/6323

Manne, R. & Kantheti, SC. 2021. Application of artificial intelligence in healthcare: Chances and challenges. Current Journal of Applied Science and Technology 40(6):78-89. https://doi.org/10.9734/ cjast/2021/v40i631320

Marais, IE. 2022. Institutionalisation of academic integrity: Experiences at a distance education university in South Africa during COVID-19. Critical Studies in Teaching and Learning 10(2):57-79. https://doi. org/10.14426/cristal.v10i2.585

Mboti, N. 2023. Artificial general intelligence, sentience, and replaceability of humans: Hype or not? A heated theoretical exchange between two University of the Free State professors. 1st Foundational Digital Capabilities Research Seminar, Tshwane, South Africa. Available at: https://www.researchgate.net/ publication/271390398_Artificial_General_Intelligence_ Concept_State_of_the_Art_and_Future_Prospects . (Accessed on 31 December 2023).

McCabe, DL. 2005. It takes a village: Academic dishonesty and educational opportunity. Liberal Education 91(3):26-31.

McCallum, S. 2023. ChatGPT banned in Italy over privacy concerns. BBC News . 1 April 2023. Available at: https://www.bbc.com/news/ technology-65139406 . (Accessed on 26 December 2023).

Medway, D., Roper, S., & Gillooly, L. 2018. Contract cheating in UK higher education: A covert investigation of essay mills. British Educational Research Journal 44:393-418. https://doi.org/10.1002/ berj.3335

Meskó, B. 2023. Prompt engineering as an important emerging skill for medical professionals: Tutorial. Journal of Medical Internet Research 25. p.e50638. 6 pages. https://doi.org/10.2196/50638

Motala, S. 2018. Achieving ‘free education’ for the poor: A realisable goal in 2018? Introduction Part III. Journal of Education 68:15-29.

Moravec, H. 1988. Mind children: The future of robot and human intelligence. Cambridge: Harvard University Press.

Mu, C. 2023. Based on natural language processing, human-computer dialogue, image recognition, and machine learning analysis whether artificial intelligence will surpass the human brain. Applied and Computational Engineering 5(1):40-47. https://doi. org/10.54254/2755-2721/5/20230526

Mwamwenda, TS. 2006. Academic integrity: South African and American university students. The Journal of Independent Teaching and Learning 1(1):34-44.

Nasrullah, N. & Al Wahyu, T. 2024. The application of Chat GPT in English Language evaluation: A systematic literature review. Futurity Education 4(3):217-235. https://doi.org/10.57125/ FED.2024.09.25.13

Neha, K. 2020. Role of artificial intelligence in education. Alochana Chakra Journal 9:305-309.

OECD (Organisation for Economic Co-operation and Development). 2023. The impact of AI on the workplace: Main findings from the OECD AI surveys of employers and workers. Available at: https://www. oecd.org/employment-outlook/2023/#ai-jobs. (Accessed on 12 December 2023).

Oliver, E. 2022a. Society 5.0 and Education 5.0 with reference to higher education. In Oliver, E. (Ed.): Global initiatives and higher education in the fourth industrial revolution, 53-68. Auckland Park: UJ Press. https://doi.org/10.36615/9781776405619-02

Oliver, WH. 2022b. Global initiatives within the 4IR and the role of higher education. In Oliver, E. (Ed.): Global initiatives and higher education in the fourth industrial revolution, 27-52. Auckland Park: UJ Press. https://doi.org/10.36615/9781776405619

Pecorari, D. 2008. Academic writing and plagiarism: A linguistic analysis. London: Bloomsbury.

Perkins, M. 2023. Academic integrity considerations of AI large language models in the post-pandemic era: ChatGPT and beyond. Journal of University Teaching & Learning Practice 20(2). 24 pages. https:// doi.org/10.53761/1.20.02.07

Perkins, M., Gezgin, UB., & Gordon, RD. 2019. Plagiarism in higher education: Classification, causes and controls. Pan-Pacific Management Science 2:1-12.

Planck. 2018. From an almost perfect universe to the best of both worlds. 17 July 2018. Available at: https://sci.esa.int/web/planck/ /60499-from-an-almost-perfect-universe-to-the-best-of both-worlds . (Accessed on 2 January 2024).

Pokrivcakova, S. 2019. Preparing teachers for the application of AI-powered technologies in foreign language education. Journal of Language and Cultural Education 7(3):135-153. https://doi. org/10.2478/jolace-2019-0025

Poola, I. 2023. Overcoming ChatGPT’s inaccuracies with pre-trained AI prompt engineering sequencing process. International Journal of Technology and Emerging Sciences 3(3):16-19.

Popenici, SAD. & Kerr, S. 2017. Exploring the impact of artificial intelligence on teaching and learning in higher education. Research and Practice in Technology Enhanced Learning 12. 22. 13 pages. https://doi.org/10.1186/s41039-017-0062-8

PWC (Price Waterhouse Coopers). 2017. No longer science fiction, AI and robotics are transforming healthcare. Available at: https://pwc. to/2weGo5v . (Accessed on 12 December 2023).

Quinlan, JR. 1986. Induction of decision trees. Machine Learning 1:81-106. https://doi.org/10.1007/BF00116251

Ragolane, M., Patel, S., & Salikram, P. 2024. AI versus human graders: Assessing the role of large language models in higher education. Asian Journal of Education and Social Studies 50(10):244-263. https://doi.org/10.9734/ajess/2024/v50i101616

Robayo-Pinzon, O., Rojas-Berrio, S., Rincon-Novoa, J., & Ramirez-Barrera, A. 2023. Artificial intelligence and the value co-creation process in higher education institutions. International Journal of Human-Computer Interaction . 17 pages. https://doi.org/10.1080/10 447318.2023.2259722

Roberts, J. 2016. Thinking machines: The search for artificial intelligence. Distillations Magazine 2(2):14-23.

Roe, J. & Perkins, M. 2022. What are automated paraphrasing tools and how do we address them? A review of a growing threat to academic integrity. International Journal for Educational Integrity 18. 15. 10 pages. https://doi.org/10.1007/s40979-022-00109-w

Roose, K. 2023. A.I. poses ‘risk of extinction,’ industry leaders warn. The New York Times . 30 May 2023. Available at: https://www.nytimes. com/2023/05/30/technology/ai-threat-warning.html . (Accessed on 7 January 2024).

Rothman, J. 2023. Why the godfather of A.I. fears what he’s built. The New Yorker . 13 November 2023. Available at: https://www.newyorker. com/magazine/2023/11/20/geoffrey-hinton-profile-ai . (Accessed on 5 December 2023).

Russ, H., Soffel, MH., Kasai, MA., & Börner, G. 1997. Age of the universe: Influence of the inhomogeneities on the global expansion factor. Physical Review D56(4):2044-2050. https://doi.org/10.1103/ PhysRevD.56.2044

Russell, SJ. & Norvig, P. 2021. Artificial intelligence a modern approach. 4th ed. London: Pearson.

Rutgers Business School. 2014. New academic integrity award will honor research of retiring Professor Donald L. McCabe. Rutgers Newark. 20 April 2014. Available at: https://www.newark.rutgers.edu/ news/new-academic-integrity-award-will-honor-research retiring-professor-donald-l-mccabe . (Accessed on 7 January 2024).

Sadler, DR. 1989. Formative assessment and the design of instructional systems. Instructional Science 18:119-144. https://doi.org/10.1007/ BF00117714

Samoili, S., Cobo, ML., Gómez, E., De Prato, G., Martínez-Plumed, F., & Delipetrev, B., 2020. AI watch. Defining artificial intelligence: Towards an operational definition and taxonomy of artificial intelligence. JRC Technical Report. Luxembourg: Publications Office of the European Union.

Sharma, RC., Kawachi, P., & Bozkurt, A. 2019. The landscape of artificial intelligence in open, online and distance education: Promises and concerns. Asian Journal of Distance Education 14(20):1-2.

Shortliffe, T. & Davis, R. 1975. Some considerations for the implementation of knowledge-based expert systems. ACM SIGART Bulletin (55):9 12. https://doi.org/10.1145/1045253.1045254

Stevens, TN. 2013. Promoting a culture of integrity: A study of faculty and student perceptions of academic dishonesty at a large public Midwestern university. PhD thesis. University of Missouri-St. Louis. Available at: https://core.ac.uk/download/pdf/217322415. pdf . (Accessed on 5 January 2024).

Stone, A. 2023. Student perceptions of academic integrity: A qualitative study of understanding, consequences, and impact. Journal of Academic Ethics 21:357-375. https://doi.org/10.1007/s10805-022 09461-5

Sutherland, SW. 2008. Plagiarism, the internet, and student learning: Improving academic integrity. New York: Routledge.

Takyar, A. 2022. AI use cases & applications across major industries. LeewayHertz-Software Development Company Publications. Available at: https://www.leewayhertz.com/ai-use-cases-and applications/ . (Accessed on 12 December 2023).

Thiebes, S., Lins, S., & Sunyaev, A. 2021. Trustworthy artificial intelligence. Electronic Markets 31:447-464. https://doi.org/10.1007/s12525 020-00441-4

Thomas, A. & De Bruin, GP. 2015. Plagiarism in South African management journals. South African Journal of Science 111(1/2). 3 pages. https:// doi.org/10.17159/sajs.2015/20140017

Timenko, M. 2021. 21st century skills in school education in the United Kingdom. Тhe Sources of Pedagogical Skills 26:202-206. https:// doi.org/10.33989/2075-146x.2020.26.227654

Trilling, B. & Fadel, C. 2009. 21st century skills: Learning for life in our times. San Francisco: Jossey-Bass.

Trott, S., Jones, C., Chang, T., Michaelov, J., & Bergen, B. 2023. Do large language models know what humans know? Cognitive Science 47(7). p.e13309. 21 pages. https://doi.org/10.1111/cogs.13309

Turing, AM. 1950. Computing machinery and intelligence. Journal of Mind 59(236):433-460. https://doi.org/10.1093/mind/LIX.236.433

Vandana, KL. & Nagaveni, NB. 2019. University grants commission (promotion of academic integrity and prevention of plagiarism in higher educational institutions) regulations, 2018. CODS-Journal of Dentistry 9(2):iv-vi.

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, AN., Kaiser, Ł., & Polosukhin, I. 2017. Attention is all you need. In Guyon, I., Von Luxburg, U., Bengio, S., Wallach, H., Fergus, R., Vishwanathan, S., & Garnett, R. (Eds.): Advances in neural information processing systems 30. NeurIPS Proceedings. Available at: https://proceedings.neurips.cc/paper_files/paper/2017. (Accessed on 11 January 2024).

Venugopal, R. & Mamatha, V. 2023. Impact of artificial intelligence (AI) on teaching and learning in India’s higher education sector. IOSR Journal of Research & Method in Education 13(5). 6 pages.

Wahle, JP., Ruas, T., Kirstein, F., & Gipp, B. 2022. How large language models are transforming machine-paraphrased plagiarism. In Goldberg, Y., Kozareva, Z., & Zhang, Y. (Eds.): Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP) in Abu Dhabi, UAE, 952-963. Available at: https://aclanthology.org/2022.emnlp-main.62/ . (Accessed on 25 January 2024). https://doi.org/10.18653/v1/2022.emnlp-main.62

Walker J. 2010. Measuring plagiarism: Researching what students do, not what they say they do. Studies of Higher Education 35(1):41-59. https://doi.org/10.1080/03075070902912994

WEF (World Economic Forum). 2016. Future of work: Ten 21st-century skills every student needs. 10 May 2016. Available at: https:// www.weforum.org/agenda/2016/03/21st-century-skills-future jobs-students/ . (Accessed on 9 January 2024).

Whitby, B. 2008. Artificial intelligence: A beginner’s guide. Oxford: One World.

William, S. 1993. On language: Words left out in the cold. The New York Times. 14 February 1993. Available at: https://query.nytimes.com/gst/fullpage. html?res=9F0CE7DF1138F937A25751C0A965958260 . (Accessed on 25 January 2024).

Winograd, T. 1971. Procedures as a representation for data in a computer program for understanding natural language. PhD thesis, Massachusetts Institute of Technology, Cambridge. Available at: https://apps.dtic.mil/sti/pdfs/AD0721399.pdf . (Accessed on 6 January 2024).

Winter, A. 2023. The Hollywood writers AI deal sure puts a lot of trust in studios to do the right thing. Wired. 28 September 2023. Available at: https://www.wired.com/story/writers-strike-hollywood-ai protections/ . (Accessed on 25 December 2023).

Wu, Z., Ji, D., Yu, K., Zeng, X., Wu, D., & Shidujaman, M. 2021. AI creativity and the human-AI co-creation model. In Kurosu, M. (Ed.): Human-computer interaction: Theory, methods and tools, 171-190. HCII 2021. Lecture Notes in Computer Science 12762. Cham: Springer. https://doi.org/10.1007/978-3-030-78462-1_13

Yadav, P., Tudela, LAM., & Marco-Lajara, B. 2024. The role of AI in assessing and achieving the sustainable development goals (SDGs). In Irfan, M., Hussainey, K., Bukhari, SAC., & Nam, Y. (Eds.): Issues of sustainability in AI and new-age thematic investing, 1-17. Hershey: IGI Global. https://doi.org/10.4018/979-8-3693 3282-5.ch001

Yeo, MA. 2023. Academic integrity in the age of artificial intelligence (AI) authoring apps. TESOL Journal 14(3). e716. 18 pages. https://doi. org/10.1002/tesj.716

Zaphiris, P. & Ioannou, A. (Eds.). 2021. Learning and collaboration technologies: New challenges and learning experiences. 8th International Conference, LCT 2021, held as part of the 23rd HCI International Conference, HCII 2021, Virtual Event, 24-29 July 2021. Proceedings Part I (Vol. 12784). Cham: Springer Nature.

Zhang, Y. 2017. ‘Type 20: Corruption in education.’ The book of swindles: Selections from a late Ming collection. New York: Columbia University Press. https://doi.org/10.7312/zhan17862

Zhang, ZV. 2020. Engaging with automated writing evaluation (AWE) feedback on L2 writing: Student perceptions and revisions. Assessing Writing 43. 100439. https://doi.org/10.1016/j. asw.2019.100439

Introduction

AI (Artificial intelligence) has permeated the world and currently has an undeniable footprint in most areas of society. In relation to the HE (higher education) sector, AI is becoming a prominent feature. AI in HE is readily accessible through the use of ‘intelligent tutoring systems, teaching robots, learning analytics dashboards, adaptive learning systems and human-computer interfaces’ (Ishak & Jiang 2022:70). Pelletier, Robert, Muscanell, McCormack, Reeves, Arbino, and Grajek (2023:4) aptly highlight the impact that AI currently has on the globe and the need to consider it as a current trend in the HE sector. Huang, Saleh, and Liu (2021:206) similarly argue that it is indisputable that AI has infiltrated the education space. As modern science and technology progress, the advancement in AI has also progressed. Research indicates that AI, when applied to education, has resulted in positive effects which have aided toward teaching reform (Huang et al. 2021:206).

Two major areas which will be significantly impacted by AI in HE are curriculum offerings and enrolment areas (Dhawan & Batra 2020:13). Joshi, Rambola, and Churi (2021:2 of 13) further highlight that AI in education has drawn attention to the following:

• Automation: AI is beneficial in automating a number of tasks. These include assessment, classifying digital assets

and schedules, and freeing up time for educators to interact with students as AI takes care of simple teaching tasks.

through the integration of AI solutions and other IT

(information technology) initiatives such as intelligent technology.

• Delineation: AI can identify fundamental trends in education thereby directing stakeholders to make decisions about the curriculum which are topical and relevant in nature.

AI has gained traction in HE due to its ability to make a contribution by means of automating aspects of the teaching experience. Additionally, personalisation is a possibility through AI applications, as well as individualised feedback and identification of topics which require revision in class. This is made possible through the AI application detecting students’ success in learning which is informed by identifying students’ content knowledge and personal characteristics (Hinojo-Lucena, Aznar-Diaz, Caceres-Reche, & Romero-Rodriguez 2019:2). Aldosari (2020:147) explains that AI can make a contribution to education for all stakeholders: This is possible though managing the integration of a variety of types of human interaction and information and communication technology.

Amidst the pervasive presence of AI in HE, issues of the impact that AI will have on QA (quality assurance) in HE comes to the fore. QA refers to a mechanism to demonstrate that the practices in which IHEs (institutions of higher education) engage in are both professional and credible (Aluko, Krull, & Mhlanga 2022:7). Stanislav (2023:1056) cautions that AI could compromise the core of HE and its mandate to create and disseminate knowledge in society. Dhawan and Batra (2020:15) further signal the use of AI as a technology which is fragile in nature thereby exposing universities to a range of risks and challenges. In this regard these authors argue that IHEs need to be cognisant of these potential risks so that they maintain the correct control of power.

Zawacki-Richter, Marín, Bond, and Gouverneur (2019:4) similarly highlights a number of risks and implications from an ethical perspective as AI continues to become more evident in HE. From the human resource space, HE staff (lecturers, tutors, student counsellors, and staff designated to the administrative space) may feel threatened by AI applications such as chatbots, online intelligent tutors, and other expert systems arguing that these commodities will replace them. Further, AI has the potential to accelerate the capabilities of learning analytics. However, systems of this nature require significant amounts of data which may include confidential information about staff and students thereby raising concerns from a privacy and data protection point of view. Bearman, Ryan, and Ajjawi (2023:370) state that AI is not just a matter for technological innovation but also represents a fundamental change in the relation between HE and broader socioeconomic interests. At this time of accelerated change, where the social shifts are as significant as the technological ones, IHEs need to set strong policies and research agendas that attend to AI and take account of ethical implications (Bearman et al. 2023:370).

In 2020, Aldosari (2020:147) has highlighted expectancies about what life would be like in 2024 with specific reference to educational technology development. These refer to pedagogy, learning approaches and models, and approaches to make these easier. With reference to the expectancies Aldosari (2020:147) highlights that the realities in 2024 would match these expectancies with AI being a notable technological development. In this context, the question arises as to whether institutions are ready for this type of unprecedented change. As such it becomes incumbent for IHEs to interrogate the use of existing technology as well as become deliberate about the implementation of new technologies, which will create the opportunity for increased teaching and learning opportunities which are flexible in nature (Pelletier et al. 2023:10). This must be considered within the context of leveraging new technologies whilst assuring quality.

This chapter serves as vantage point to establish the use of AI in HE, consider the opportunities and challenges for using AI in HE, and most importantly establish a link between AI and QA.

Approach Used for this Study

An approach adopted by Grassini (2023) is followed in this study, starting with the main research question of this study which is: What is the link between AI and QA? This research question was explored primarily through published literature. Google scholar was used as the primary search engine using the search strings ‘artificial intelligence,’ ‘higher education,’ and ‘challenges and opportunities.’ The literature was surveyed in a nonsystematic manner.

The years from 2019 to 2023 were included to represent a five year period, which is current and relevant in nature. The initial articles were surveyed and starting points with the snowball method were used to find additional articles. Additional articles were, for example, identified from the reference lists of the selected articles. This chapter uses the articles as informants for exploring the main research question. In the next section the definition of AI is explored to gain a better understanding of the term.

What is Artificial Intelligence?

‘AI’ is a term which dates back to the 1950s. It arose from the work of scientists who explored the idea of AI. Seminal developments include the creation of an AI programme referred to as ELIZA as noted by King (2023:2). This programme was made to copy human speech. As the time periods advanced so did AI developments parallel to the time periods. Later years saw the development of more advanced chatbots and other AI applications which had the potential to respond to and comprehend requests which were complex in nature (King 2023:2; Crompton & Burke 2023:2).

Buchanan (2005:54) explains that AI goes beyond robots and should be understood from the perspective of grasping the nature of intellectual thinking and behaviour using computers as experimental devices. During its determining years, AI was influenced by a number of disciplines (Buchanan 2005:56) including engineering, biology, experimental psychology, philosophy, mathematics, statistics, and linguistics. Buchanan (2005:56) highlights that a seminal paper published in 1950 by Turing in the philosophy journal Mind, acts as a turning point in the history of AI. Turing’s paper sets a seminal discourse about the prospect of programming a computer to behave intelligently. Initial programmes in AI were limited by the technology of the time, nonetheless they demonstrated an impressive ability in answering problems that only people had been able to previously answer (Buchanan 2005:57).

There is a plethora of definitions for the term ‘AI.’ Kelly, Kaye, and Oviedo-Trespalacios (2023:2 of 33) highlight that providing a definition for AI at an academic, government, and community level poses a challenge. AI is a significantly contested concept and there has not been much consensus regarding its definition in the different fields in which it is used (Kelly et al. 2023:2 of 33). For the purposes of this chapter a few selected definitions are discussed below.

Kengam (2020) describes AI as a computer-based technology which has the ability to offer personalised, adaptive, and insightful educational experiences. The description of AI as put forward by Ocana-Fernandez, Valenzuel-Fernandez, & Garro-Aburto (2019:557) assume AI to be related to the design of intelligent systems which display the characteristics associated with human intelligence. Jain & Jain (2019:145) claim that AI is the impersonation of human comprehension and virtual decision making by robots and machines.

According to Chen, Chen, and Lin (2020:75265), AI is associated with a supercomputer which has in nature a massive processing potential. Functions which this supercomputer can execute, include adaptive behaviour whereby the inclusion of sensors and a range of capabilities provide the AI ability to mimic humanlike thinking. Hinojo-Lucena et al. (2019:2) opines that AI is a technology which is aimed at the construction of computer systems. These computer systems display smart and adaptive behaviours with the capability to acquire information from their environments mimicking what humans can do. The ability of computers to behave in a similar manner as the human brain for the improvement of the world is the description brought forward by Dhawan and Batra (2020:12). Aldosari (2020:145) outlines that AI is a distinct and current application of information systems which has a vested interest in unpacking and understanding human intelligence and replications to create the current generation of smart computers. These smart computers require a significant degree of ability to accomplish tasks which require inference, deduction, and perception.

While the latter cited authors offer different descriptions of AI the overriding commonalties which emanate from these definitions is that AI denotes the use of computers to simulate human thinking. Figure 3.1 helps to demonstrate this.

According to Kaul, Enslin, and Gross (2020:808), there are four different subfields of AI which can be distinguished:

While it is important to delineate the different subfields of AI, this chapter will not focus on a specific subfield but instead on AI in its general context.

How AI is used in Higher Education

To establish the manner in which AI is used in HE the work of Crompton and Burke (2023) is referred to. These authors have conducted a systematic review of AI in HE from 2016 to 2022. A summary of their findings highlight five areas for the use of AI in HE, which will be discussed in the sections below.

Opportunities for the Use of AI in HE

The literature (cf. Grassini 2023; Pelletier et al. 2023; Dhawan & Batra 2023) presents an array of opportunities for the use of AI in HE. Grassini (2023:3 of 13) highlights that Gen-AI (generative artificial intelligence), for example ChatGPT (chat generative pre-trained transformer), offers the advantage of creating lesson plans and presentations which engage recipients and a range of educational resources. New and innovative teaching methods are also an added opportunity as educators will have more support available through AI thereby creating time and space for more engaging teaching resources and environments. Additionally, Kengam (2020), as well as Chatterjee and Bhattacharjee (2020:3444) present the aspect of customisation in the learning space. In this regard learning environments can be tailored to suit the individual learning needs of students. Regarding assessment, automated grading alleviates the workload of the educator thereby creating space for them to utilise the saved time on the development of students.

ITSs, which facilitate the process of using automated tutors to teach different disciplines and human NLP (natural language processing) systems are examples of AI in HE which have had positive impacts. Ocana-Fernandez et al. (2019:562) highlight that these systems create an opportunity for diverse classrooms with the potential to individualise the learning experience. Pelletier et al. (2023:10) further discuss the impact of AI on HE. In this regard, AI can assist the educator with the creation of teaching content and automated grading. This creates the opportunity for educators to have additional time to focus on higher order pedagogical tasks thereby creating new knowledge for their students. The use of avatars and the metaverse for example, also creates engaging opportunities and personalisation for students. With the help of AI universities around the world are enrolling an increased number of students due to an increased flexibility and speed (Jain & Jain 2019:146).

Customisation of the learning experience as discussed by Dhawan and Batra (2020:17) is possible through AI, where a ‘one-size-fits-all’ approach can be altered to suit the needs of each individual student. Owing to the range of factors that influence learning, some of which include the knowledge that the student brings to the learning encounter, the rate at which students learn and absorb knowledge, and cultural views, AI can serve as a mechanism to accommodate different learning styles. Dhawan and Batra (2020:17) provide the example of emotion AI, which makes it possible to study behaviour which is non-verbal in nature. Emotion AI has the ability to respond to cues such as fear, happiness, and anger and deliver lessons and establish how engaged students are in the lesson. AI can further equip students with ‘tech-savvy’ skills necessary for the working environment.

Virtual assistants can be of great benefit in terms of responding to student queries. Additionally, virtual assistants can help with the enrolment, monitoring drop-out rates and retention, and prompting university staff to intervene accordingly. From a research perspective AI can be useful in managing structured and unstructured data, processing data, and suggesting new models with the potential to revise research practices in disciplines which are traditionally defined. AI can additionally assist administrators in education regarding decision-making in terms of the development of courses, pedagogical design, and transformation in the academic space (Ouyang, Zheng, & Jiao 2022:7895).

Kuleto, Ilić, Dumangiu, Ranković, Martins, Păun, & Mihoreanu (2021:7 of 16) highlight that AI can offer IHEs a number of benefits. Key processes such as student enrolment, curriculum planning, the creation and offering of personalised lessons, and mitigating high dropout numbers can be assisted via AI. As these processes can be challenging to manage, IHEs may consider using AI to manage them. Specifically learning analytics can unearth student data for the management of these processes. Chatbots and virtual tutor assistants can also assist. Repetitive tasks which are time consuming in nature can also be supported by AI, thereby offering an improved learning experience (Kuleto et al. 2021:7 of 16). AI is further beneficial in the curriculum space enabling personalised learning experiences through the use of virtual reality and robotics, among others.

AI has the ability to provide instruction, guidance, and feedback which are prompt in nature. In this manner time delays are avoided and unlike humans who may become tired due to large volumes, this will not present a challenge to AI (Crompton & Burke 2023:17). Table 3.1 below summarises the opportunities for the use of AI in HE.

Challenges for the Use of AI in Higher Education

The progression in AI has raised ‘deliberations around academic integrity, accuracy, fairness and equity’ (Pelletier et al. 2023:10) in the HE space. Further concern has been highlighted in terms of whether AI supplies output which is accurate and unbiased in nature. Outputs for example could refer to automated grading which can be questioned in terms of its accuracy. Developments in Gen-AI (e.g., ChatGP) are designed to generate answers for students. This automated generation of answers threatens the potential for students to meet learning aims of the module as they do not engage in the formulation of their own answers (Stanislav 2023:1057) and higher order thinking skills.

Concerns of ethics is also associated with the use of AI. In the HE space, the behaviour of all stakeholders is regulated and underpinned by ethics. This includes teaching, development of assessments, marking, and the manner in which students and staff interact and communicate with others. AI could pose a challenge to ethical behaviour in terms of thinking on behalf of the human being. Engaging in critical thinking and the ability for students to be innovative has a distinct link to economic and social effects as well as the ability of IHEs to compete with their counterparts (Stanislav 2023:1057).

Huang et al. (2021:212) highlight that in developing countries the use of new technologies can exacerbate a division between people. Basic infrastructure and technology must be in place for AI to provide equal opportunities for all to mitigate a situation where the digital divide is widened. Issues of safety and ethics also become a challenge in the AI space when data are collected, used, and shared. AI applications require of educators to master digital skills as the programmes which offer teaching and learning are built on applications which do require a range of digital literacy skills. The deterioration of students’ social and communication skills also poses a challenge in the AI environment. If students primarily communicate via machines, a lesser requirement of human communication is needed. As such, AI therefore poses a challenge in terms of communication skills (Huang et al. 2021:213). Hutson, Jeevanjee, Vander Graaf, Lively, Weber, Weir, Arnone, Carnes, Vosevich, Plate, and Leary (2022:3969) highlights alarming ethical concerns for the use of AI in education, with bias and privacy noted as the most prominent AI concerns.

Ouyang et al. (2022:7908-7909) narrate that AI faces a number of challenges in the education space. Questions regarding how to meet the learning needs of students, what to provide to them, and the mechanism to provide them with the correct degree of agency are some of the challenges facing AI. A range of systematic reviews indicate a common challenge in terms of the use of AI in education. This common challenge depicts a lack of a link between AI practices and implementation, and theoretical frames of reference (Ouyang et al. 2022:7908-7909).

A summary of the challenges for AI in HE is highlighted in Table 3.2.

Establishing the Link between AI and QA

To establish a possible link between AI and QA the work of Mohee (2019) is referred to. Mohee (2019) highlights that QA within the HE space needs to factor in the following: Programme design, learner support systems, materials development, student assessment, infrastructure, and facilities. The author of this chapter uses Mohee’s factors of consideration and relates them to the opportunities and challenges of AI (as highlighted in Tables

3.1 and 3.2). Both these tables are used as a point of reference to establish the possible link between AI and QA.

Table 3.3 presents the key QA factors as highlighted by Mohee (2019) and the possible link they have to the key opportunities as highlighted in Table3. 1.

Table 3.4:Establishing the link between AI challenges and quality assurance factors

An evaluation of Tables 3.3 and 3.4 highlights that each of Mohee’s factors (i.e., programme design, student support systems, materials development, student assessment, infrastructure, and facilities) features across the cited AI opportunities and challenges. This implies that the opportunities presented by AI can have a positive impact on QA.

Lesson plans, presentations, educational resources, and innovative teaching methods as cited by Grassini (2023) link

to programme design, student support systems, materials development, and student assessment. In this regard AI opportunities could have a positive impact on QA in terms of programme design, student support systems, materials development, and student assessment. Kengam (2020), as well as Chatterjee and Batacharjee (2020:3444) highlight customisation, individual learning needs, automated grading, and additional instructional time as AI opportunities. These in turn link to programme design, student support systems, and student assessment QA factors implying a positive impact in this regard. Individual learning experiences (Ocana-Fernandez et al. 2019:555) have a direct link to student support systems, while the creation of teaching content, automated grading, added instructional time, and personalised learning (Pelletier et al. 2023:10) can contribute positively to the QA factors of programme design, student support systems and materials development. Increased student enrolment (Jain & Jain 2019:146) is a positive QA factor made possible through infrastructure and facilities.

The customisation of the learning experience, accommodation of different learning styles, and equipping students with ‘tech-savvy’ skills link to the QA factor of student support systems, indicating a positive connection in this regard. Virtual assistants by way of responding to student queries, assisting with student enrolment, monitoring dropout rates and retention, managing research, and assisting in decision making through the development of courses as well as pedagogical design and transformation (Ouyang et al 2022:7895), have a direct link and positive contribution to the QA factors of infrastructure and facilities, programme design, learner support systems, and materials development.

Student enrolment, curriculum planning, personalised lessons/learning experience, mitigating a high dropout rate and the use of chatbots and virtual tutor assistants (Kuleto et al. 2021:7 of 16) on the other hand link to and have a positive impact on the QA factors of infrastructure and facilities, programme design, student support systems, and materials development. Finally, instructional guidance, feedback promptly provided to students, and responding to queries on a mass scale (Crompton & Burke 2023:17) positively influence the QA factors of student support systems, student assessment, and infrastructure and facilities.

Evaluating the link between Mohee’s QA factors and the key AI challenges it appears that academic integrity, accuracy, fairness, and equity (Pelletier et al. 2023:10) could negatively influence student assessment in the HE space. Further accurate unbiased output (automated grading), the generation of automated answers, a lack of higher order thinking skills, concerns regarding ethics, thinking on behalf of humans, and a lack of thinking from students (Stanislav 2023:1057) could impact IHEs’ competitiveness and have a negative effect on the QA factors of student assessment, programme design, student support systems, materials development, and infrastructure and facilities. A widening of the digital divide in developing countries, the requirement for educators to have digital skills, and a deterioration of student communication skills (Huang et al. 2021:212) are linked to the QA factors of infrastructure, facilities, and programme design. This implies that if there is not a proper infrastructure and facilities in place AI could widen the digital divide in developing countries, additionally a deterioration of student communication skills could negatively impact programme design.

Hutson et al. (2022) raise the challenge of ethics with specific reference to privacy and bias. If IHEs do not have competent infrastructures and facilities in place, ethics could be compromised.

Lastly, an AI application needs to make decisions in terms of what to provide to students, how to provide resources to them, and providing them with the appropriate degree of agency. If these decisions are not appropriately taken QA factors of programme design could be negatively impacted.

The discussions above provide an evaluation of the link between AI and QA from both a positive and negative lens. In terms of impact the positive links between AI opportunities and QA as highlighted above could then positively influence the QA factors mentioned. On the other hand, the negative links between AI challenges and QA factors, as highlighted could impact negatively on QA factors as demonstrated in the discussion above.

What this implies is that IHEs need to carefully look at the challenges and opportunities that AI offers. Where AI opportunities can improve QA factors these must be carefully implemented. Conversely the challenges that AI presents should be carefully studied and strategies should be developed to mitigate these challenges from impacted negatively on QA factors.

Conclusion

This chapter aimed at establishing a link between AI and QA. To achieve this an introduction to AI was first offered. This was followed by attempting to define AI albeit the literature highlighted that there are various definitions for AI. Ultimately this chapter offered a definition for AI which refers to the simulation of human thinking through the use of computers. Next the use of AI in HE was discussed. This highlighted assessment (evaluation), AI assistance, intelligent tutoring systems, and managing student learning as key uses for AI in HE.

Following this the opportunities and challenges for the use of AI were presented. Key opportunities demonstrated are innovative teaching methods, customisation, automated grading, added instructional time, personalised learning, virtual assistants, and increased enrolment, among others.

The challenges for the use of AI in HE followed next which highlighted some of the following: Academic integrity, generation of automated answers, concerns regarding ethics, widening of the digital divide in developing countries, issues of privacy and bias, and a lack of connection between AI and theoretical frames of reference, to mention a few.

Lastly, a possible link between AI and QA factors was established. Quality assurance factors featured across both the cited challenges and opportunities. From this discussion it was highlighted that the opportunities AI presents link to QA factors and can improve QA factors. On the other hand, there was also a link between the challenges provided and QA factors. However, it came to light that for QA factors not to be compromised the challenges presented by AI must be carefully mitigated.

As universities continue to traverse the changing HE landscape and specifically the pervasive nature of AI in HE, there must be constant reflection about strategies to best leverage the influx of persistent change associated with the HE landscape.

References

Aldosari, SA. 2020. The future of higher education in the light of artificial intelligence transformations. International Journal of Higher Education 9(3):145-151. https://doi.org/10.5430/ijhe.v9n3p145

Aluko, R., Krull, G., & Mhlanga, E. 2022. Shaping open, distance and e-learning in post school education and training: A call for a revised agenda. Perspectives in Education 40(1):1-17. https://doi. org/10.18820/2519593X/pie.v40.i1.1

Bearman, M., Ryan, J., & Ajjawi, R. 2023. Discourses of artificial intelligence in higher education: A critical literature review. Higher Education 86:369-385. https://doi.org/10.1007/s10734 022-00937-2

Buchanan, BG. 2005. A (very) brief history of artificial intelligence. AI Magazine 26(4):53-60. https://doi.org/10.1609/aimag.v26i4.1848

Chatterjee, S. & Bhattacharjee, KK. 2020. Adoption of artificial intelligence in higher education: A quantitative analysis using structural equation modelling. Education and Information Technologies 25:3443-3463. https://doi.org/10.1007/s10639-020-10159-7

Chen, L., Chen, P., & Lin, Z. 2020. Artificial intelligence in education: A review. IEEE Access 20:75264-75278. https://doi.org/10.1109/ ACCESS.2020.2988510

Crompton, H., & Burke, D. 2023. Artificial intelligence in higher education: The state of the field. International Journal of Educational Technology in Higher Education 20: 1-22. https://doi.org/10.1186/ s41239-023-00392-8

Dhawan, S. & Batra, G. 2020. Artificial intelligence in higher education: Promises, perils, and perspective. OJAZ Expanding Knowledge Horizon 11:11-22.

Grassini, S. 2023. Shaping the future of education: Exploring the potential and consequences of AI and ChatGPT in educational settings. Education Sciences 13(7). 692. 13 pages. http://dx.doi.org/10.3390/ educsci13070692

Huang, J., Saleh, S., & Liu, Y. 2021. A review on artificial intelligence in education. Academic Journal of Interdisciplinary Studies 10(3):206 217. https://doi.org/10.36941/ajis-2021-0077

Hinojo-Lucena, FJ., Aznar-Diaz, I., Caceres-Reche, MP., & Romero-Rodriguez, JM. 2019. Artificial intelligence in higher education: A bibliometric study on its impact in the scientific literature. Education Sciences 9(1):2-9. https://doi.org/10.3390/ educsci9010051

Hutson, J., Jeevanjee, T., Vander Graaf, V., Lively, J., Weber, J., Weir, G., Arnone, K., Carnes, G., Vosevich, K., Plate, D., & Leary, M. 2022. Artificial intelligence and the disruption of higher education: Strategies for integrations across disciplines. Creative Education 13(12):3953-3980. https://doi.org/10.4236/ce.2022.1312253

Ishak, NA. & Jiang, CX. 2022. Research on the influence of artificial intelligence technology on the formulation of educational strategies. International Journal of Intelligent Systems and Applications in Engineering 10(2s):70-75.

Jain, S. & Jain, R. 2019. Role of artificial intelligence in higher education – an empirical investigation. International Journal of Research and Analytical Views 6(2):144-150.

Joshi, S., Rambola, RK., & Churi, P. 2021. Evaluating artificial intelligence in education for next generation. Journal of Physics: Conference Series . 1714 012039. 13 pages. https://doi.org/10.1088/1742 6596/1714/1/012039

Kaul, V., Enslin, S., & Gross, SA. 2020. History of artificial intelligence. Medicine Gastrointestinal Endoscopy 92(4):807-812. https://doi. org/10.1016/j.gie.2020.06.040

Kelly, S., Kaye, S-A., & Oviedo-Trespalacios, O. 2023. What factors contribute to the acceptance of artificial intelligence? A systematic review. Telematics and Informatics 77. 101925. 33 pages. https://doi.org/10.1016/j.tele.2022.101925

Kengam, J. 2020. Artificial intelligence in education. Available at: https:// www.researchgate.net/publication/347448363_ARTIFICIAL_ INTELLIGENCE_IN_EDUCATION . (Accessed on 23 January 2024).

King, MR. 2023. A conversation on artificial intelligence, chatbots, and plagiarism in higher education. Cellular and Molecular Bioengineering 16:1-2. https://doi.org/10.1007/s12195-022 00754-8

Kuleto, V., Ilić, M., Dumangiu, M., Ranković, M., Martins, OMD., Păun, D., & Mihoreanu, L. 2021. Exploring opportunities and challenges of artificial intelligence and machine learning in higher education institutions. Sustainability 13(18). 10424. 16 pages. https://doi. org/10.3390/su131810424

Mohee, R. 2019. Quality assurance: Good practices in ODL in Sub-Saharan Africa. Available at: http://hdl.handle.net/11599/3132. (Accessed on 14 April 2023).

Ocana-Fernandez, Y., Valenzuel-Fernandez, L., & Garro-Aburto, L. 2019. Artificial intelligence and its implications in higher education. Propitos y Representaciones 7(2):536-568. https://doi. org/10.20511/pyr2019.v7n2.274

Ouyang, F., Zheng, L., & Jiao, P. 2022. Artificial intelligence in online higher education: A systematic review of empirical research from 2011 to 2020. Education and Information Technologies 27:7893 7925. https://doi.org/10.1007/s10639-022-10925-9

Pelletier, K., Robert, J., Muscanell, N., McCormack, M., Reeves, J., Arbino, N., & Grajek, S. 2023. EDUCAUSE Horizon Report: Teaching and learning edition. Boulder: EDUCAUSE.

Stanislav, I. 2023. The dark side of artificial intelligence in higher education. The Service Industries Journal 43(15/16):1055-1082. https://doi.org/10.1080/02642069.2023.2258799

Zawacki-Richter, O., Marín, VI., Bond, M., & Gouverneur, F. 2019. Systematic review of research on artificial intelligence applications in higher education – where are the educators? International Journal of Educational Technology in Higher Education 16(1):1-27. https://doi.org/10.1186/s41239-019-0171-0

Introduction

The emergence and ongoing development of the information era have given rise to various ethical dilemmas that are inherent to the essence of information. IE (information ethics) is an interdisciplinary area that incorporates disciplines such as philosophy, computer science, sociology, law, and others, as highlighted by scholars like Quin (2011) and Floridi (2019). Although the information age is frequently regarded as a contemporary occurrence, this perspective fails to acknowledge its underlying historical origins. The origins of this age can be attributed to previous advancements in communication and Its (information technologies). The information age is currently recognised as a dynamic and ongoing era characterised by swift technological developments that are continuously transforming various aspects of society, such as communication, education, healthcare, and entertainment.

IE is a discipline that combines ethical theories with practical considerations about technology. It acknowledges the intricate and diverse character of ethical dilemmas in the digital world. The subject matter encompasses inquiries on data privacy, cybersecurity, digital rights, and the ethical utilisation of AI (artificial intelligence) and ML (machine learning). With the advancement and widespread use of technology, ethical questions become more complex and important.

Furthermore, IE is dynamic, constantly adjusting to address novel issues and inquiries presented by evolving technologies. For example, the emergence of social media platforms has ignited issues on misinformation, digital identity, and online conduct, while progress in AI and robotics is stimulating conversations about machine ethics, job displacement, and the prospects of human-machine interactions in the future.

As we traverse this changing environment, it is crucial to comprehend and tackle the ethical consequences of our digital existence. In this chapter the findings of a review of recent literature allude to the need to address issues of the privacy and security issues related to data collection, digital identity and authenticity, accessibility and digital inclusivity, legal frameworks, surveillance, and the freedom of expression. The study delves into how AI innovation influences new nuances in IE. The ultimate goal is to establish a world that is not only technologically sophisticated but also firmly rooted in ethical values.

Contribution of Information Science and Informatics to Information

What is Information Ethics?

IE, in it its simplest form, refers to the use of information in a way that it does not cause harm. As technology evolves, new ethical challenges and dilemmas arise (Moor 1985; 2020; Burgess & Knox 2019). As a subset of information science and being part of applied ethics, IE examines ethical issues associated with the creation, dissemination, and use of information. It is particularly relevant to the informatics discipline that deals with the management and analysis of information.

On a macro level Zhou, Chen, Berry, Reed, Zhang, and Savage (2020:3010) share that ethics as a branch of philosophy refers to the systematising, defending, and recommending concepts of right and wrong conduct, where rights, obligations, benefits, and fairness principles are explained and deliberated in models and frameworks.

The foundations of IE are instituted in its key principles and constructs. Figure 4.1 offers a summary of the overlapping constructs included in theories and research that will be discussed in greater detail this chapter.

These foundational principles guide the development of ethical frameworks, policies, and practices in the rapidly evolving landscape of IT and data management. Adhering to these principles helps to ensure that information is used responsibly, ethically, and in ways that benefit individuals and society. IE is at the heart of AI and the ethics of AI (Bester & Fisher 2020:2). These ethics are underpinned by the philosophy of information.

In the Beginning: The Philosophy of Information and Information Ethics

PI (the philosophy of information) encompasses aspects of philosophy, computer science, information science, cognitive science, and communication studies. It addresses fundamental questions related to the nature of information, its role in the universe, and its impact on knowledge, reality, and society (Bynum 2010:420). As an interdisciplinary field, it explores the nature, properties, and implications of information. Floridi (1999:41) defines PI as a field of philosophy, more specifically looking at critical research into the basic principles of information and the conceptual nature thereof. This field includes the dynamics of information, its uses and applications theory, and computational methodologies (Floridi 2013:38; 2015:42). IE, in turn, is a branch of PI. New developments, such as AI and Gen-AI (generative artificial intelligence), have a profound impact on moral decision-making in our daily lives.

Moral tenets and theories can be complex, and when revisiting the discourses of seminal authors it lays a foundation to meaningfully expand on the ever increasing moral and ethical issues of AI. Floridi (2015:197) stresses the importance to find a balance in practices and experiences in daily lives and uses the metaphor of a tree depending on its roots for the sustainable growth of new branches. Seminal theorists of IE are particularly found in mathematics (Shannon 1948), informatics (Turing 1950), and information science (cf. Mason 1986; Capurro, Eldred, & Nagel 2013).

Finding the Roots of Information Ethics

On the challenges encountered when applying human-centred philosophy of morals and ethics in ICTs (information and communication technologies), Birrer (1999:16) laments: ‘It is by no means accidental that most of the discussion about values in technology focuses on normative components that are considered undesirable: Technocracy, dehumanization, inequality, manipulation, loss of control.’ Values and norms are humanistic and not technological constructs (Moor 2006:18). As such, applying moral and ethical theories to information and ICTs is complex.

Expanding upon the first ideas about the essence and intricacies of AI, we arrived at the overarching question: Can computers, like humans be held responsible for their acts? One of the notable futurists and mathematicians was Alan Turing. In the early fifties of the previous century he was a pioneering mathematician and computer scientist, who introduced the Turing Test to assess the intelligence of machines. In 1950 his book, Computing machinery and intelligence (Turing 1950) posed and addressed the questions on whether computers can ‘think.’ The Turing Test involves a human judge engaging in natural language conversations with both a human and a machine, without knowing which is which. If the judge cannot reliably distinguish between the human and the machine based on the conversation, then, according to Turing, we could say that the machine is demonstrating a form of intelligence indistinguishable from that of a human.

The Turing Test remains a widely discussed concept in AI and philosophy, and it has sparked debates about the nature of consciousness, intelligence, and the potential capabilities of machines. The Turing Test is once again in the limelight, this time on the development of LLMs (large language models) used in Gen-AI.

Early Philosophers and Pioneers of the Information Ethics Theory

One of the first and influential models in IE is associated with the work of Norbert Wiener. In 1948, Wiener, a mathematician and philosopher pioneered research in cybernetics, a precursor to IE. His work laid the foundation for discussions on the ethical implications of ICT. While Wiener’s work may not fit the contemporary definition of a comprehensive IE model, it has set the stage for ethical reflections on the impact of technology on individuals and society, addressing issues such as control, communication, and the responsibilities associated with the use of information in automated systems.

IE Models and Frameworks

Prominent frameworks that have been informing IE research, teachings, and practices include, among others, the 2013 privacy framework by the OECD (Organisation for Economic Cooperation and Development) FIP (fair information practice) (OECD 2013), Floridi’s IE frameworks, the PAPA (privacy, accuracy, property, and accessibility) framework, and Tavani’s informational privacy framework. This chapter provides an overview of IE models and frameworks and discusses whether the capability of existing frameworks is comprehensive enough to provide the ethical guardrails offered to accommodate the ethical considerations prevalent in a fast-changing world. IE typically considers examining the ethical implications and considerations associated with disruptive innovations around the creation, use, and management of information within virtual and immersive environments. The question now is to what extent exiting frameworks are still relevant to underpin the vast new developments.

While the frameworks mentioned above, provide a useful foundation for evaluating ethical issues in technology, including aspects of Gen-AI, there may be potential shortcomings requiring further adaptation to address the unique challenges posed by new developments. Han (2022:1 of 11) reports that the oldest and most referred to terms applied to the ethical use of information in a digital environment are ‘computer’ and ‘IE.’ The finding of this study is that new aspects must be considered, such as cross-cultural aspects on a global scale, devising guidelines for content moderation, ensuring that AI systems are ethical and avoid discriminatory practices, and lastly, considering environmental impact ethics. The conclusion is that one should take the wisdom of existing frames and build a comprehensive, adaptable multidisciplinary framework on it to address ethical dilemmas presented by evolving technological advancements and changing societal norms. Such a framework should encourage responsible innovation by providing the ethical guardrails, and simultaneously safeguarding the rights, wellbeing, and dignity of individuals in virtual spaces.

The Ethics of AI

The recent hype around AI and Gen-AI created the illusion that AI is the ultimate panacea (Floridi 2020:2). Views are ranging from

ultimate doom to a hailing of AI as the beginning of everything. This is not realistic and authors such as Floridi (2019; 2020) as well as Sartori and Theodorou (2022:10 of 11) remind us that during the many decades of the becoming of AI, there were many wins and losses. What cannot be denied, is the rapid tempo of recent developments of LLMs and prompt engineering. The sociotechnical nature of AI and Gen-AI remains uncontested, where the development, deployment, and impact of AI are deeply intertwined with societal factors, human behaviour, and ethical considerations.

The Nature of AI and Gen-AI

AI and AI ethics are not new fields of research and study. Floridi (2020:3), as well as Sartori and Theodorou (2022:2 of 11) point to the many ‘winters’ and ‘summers’ that AI has already endured. In 1956, a group of mathematicians and computer scientists met at the Dartmouth College, New Hampshire, to discuss the emerging field of AI (Strickland 2021). It was here that John McCarthy coined the term ‘artificial intelligence,’ which he argued, ‘would explore the hypothesis that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it’ (Strickland 2021:27). Bartneck, Lütge, Wagner, and Welsh (2020:18) allude that a complete definition of AI is still in the making, but they conceptualise AI as the simulation of human intelligence processes by machines, especially computer systems. Specific applications of AI include expert systems , natural language processing , speech recognition, and machine vision.

The emergence of Gen-AI technologies expanded significantly on existing AI technologies, and opened numerous opportunities as well as future challenges. As part of ML Gen-AI uses algorithms that depend on large datasets. Current Gen-AI models and tools can produce original material, such as text, images, music, and even complex designs, based on their training data. Floridi (2024) posits that these innovations drastically change the supply and demand of information.

AI Ethics

Essentially, AI ethics is a system of moral principles and techniques intended to inform the development and responsible use of AI technology (Bartneck et al. 2020:2247). Han (2022:2 of

11) reminds us that there is no clear and agreed definition of AI, as found in other fields of ethics such as computer ethics and IE. In literature there are overlaps in terminology and often AI ethics and robotic ethics are used interchangeably (cf. Figure 3). As AI has become integral to product platforms that incorporate ethical considerations and values in the development and deployment of AI and services, organisations are starting to develop AI codes of ethics. Han (2022:3 of 11) refers to Moor (1985:266) who defines computer ethics as an analysis of the nature and social impact of computer technology and the corresponding formulation and justification of policies for the ethical use of such technology (Han 2022:3 of 11). Hauptman who was the first to use the term ‘IE’ in 1988, already addressed principles of censorship, privacy, access to information, balance in collection development, copyright, fair use, and codes of ethics (Froehlich 2000:265).

In this model, cyborg – short for cyborg organism – and cyborg ethics refer to a combination of biological and artificial components. These enhanced capabilities are often present in biomedical technologies such as pacemakers, implants, prosthetics, or other technological enhancements that merge with the biological components of the organism. Han’s model provides a contemporary view of the span and complexity of new IE.

An AI code of ethics, also called an AI value platform, is a policy statement that formally defines the role of AI as it applies to the development and wellbeing of humans. This term should not be confused with AI value-driven platforms, looking at capabilities and benefits of systems. The purpose of an AI code of ethics is to provide stakeholders with guidance when faced with an ethical decision regarding the use of AI (Ibiricu & Van der Made 2020:396). An AI value platform is generally understood as technology purposed to add value via AI. This could involve a combination of software, hardware, and services designed to provide solutions that address specific business challenges or opportunities.

It is important to note that the interpretation of the AI value platform may vary depending on the context and the specific goals of the entity using or providing such a platform. Organisations developing AI solutions are often looking for platforms that not only facilitate the technical aspects of AI but also deliver measurable values in terms of efficiency, innovation, and competitive advantage. For the latest and most accurate information, it is recommended to check specific industry announcements, publications, or product documentation.

It is quite incidental how Asimov’s science fiction short stories around 1950 referred to an ethical code where the primary law prohibits robots from engaging in any actions that cause harm to humans or from neglecting to act when harm could be prevented (Asimov 1984:9). The second law mandates that robots must comply with human commands, except when those commands conflict with the first law. The third law mandates that robots prioritise self-preservation, if it aligns with the principles outlined in the previous two laws.

AI Ethics Frameworks and Principles

AI ethics must consider the ethical implications and societal impacts of AI technologies. Building on existing IE frameworks, new theoretical frameworks have been proposed to guide discussions and decision-making in this domain. These include the IEEE (Institute of Electrical and Electronics Engineers) global initiative on ethics of autonomous and intelligent systems and the COMEST (Commission mondiale d’éthique des connaissances scientifiques et des technologies) rights-based model: The universal declaration on ethical considerations regarding artificial intelligence and autonomous systems.

AI Risks

In the dynamic world of technology, ethical risks of AI do not only feature in its use, but also in its design. Addressing challenges of biases in data, algorithms, as well as training set limitations leaves room for deeper discourse and ongoing efforts to improve data quality, enhance algorithmic fairness, ensure robustness, and consider ethical implications throughout the AI development lifecycle (cf. Figure 4.2).

AI could be prone to inaccurate information and misrepresentation. Some of these challenges include incidences of deep fake, identity theft, and privacy issues in using facial recognition, to name but a few.

AI Threats

A dynamic invention like AI invariably poses some threats to society. Much of these lies in the misuse of AI like cyberwarfare, and there is also the overarching fear of loss of human control when autonomous systems move closer to super-intelligence.

Ethics of Gen-AI

It should be noted that the technology of Gen-AI is not new. Gen-AI was introduced in the 1960s in chatbots. However, it was not until 2014, with the introduction of GANs (generative adversarial networks) – a type of ML algorithm – that Gen-AI could create convincingly authentic images, videos, and audio of real people. Gen-AI is a type of AI technology that can produce various types of content, including text, imagery, audio, and synthetic data. Gen-AI refers to a class of AI systems that can generate new content, such as images, text, or even music, that is not explicitly programmed. These systems are designed to understand and learn patterns from existing data and then use that understanding to create entirely new data that resemble the training set. Bandi, Adapa, and Kuchi (2023:1 of 60) explain that Gen-AI is geared to develop algorithms able to create synthetic data. There are several types of Gen-AI models, and one of the most notable is the class of generative models based on neural networks.

Gen-AI’s ability to generate data resembling real-life conditions, allows for content creation (Bandi et al. 2023:1 of 60). It is a further development of Web 3.0. Gen-AI has found applications in various domains, including text generation in NLP (natural language processing) models, such as GPT 3 (generative pre-trained transformer 3) and GPT 4 and data augmentation of data training sets, and can generate coherent and contextually relevant text. These models can be used for tasks like writing articles and generating creative content. Then there is the image generation, where generative models can create realistic images that resemble photographs of human faces, animals, or scenes.

While Gen-AI has shown remarkable capabilities, there are also ethical concerns, especially regarding the potential misuse of the technology, such as the creation of deepfakes for malicious purposes. As a result, ongoing research and development in the field aim to address both the positive and negative implications of Gen-AI.

Intergovernmental and other Initiatives in Support of New Dimensions in Dynamic IE Ethics Innovation

Floridi (2014:218) urges that any information society must be equipped with sustainable IE, and that this be made well known. It is here where intergovernmental organisations have a critical role to play, moreover in the time where disruptive technologies could potentially add to chaotic and unethical consequences. An example of the speed at which innovation takes place is the rapid development of the Gen-AI by OpenAI in bringing out its chatbot.

UNESCO has made great strides in addressing frontier challenges in IE, the ethics of AI, and the ethics of neurotechnology. In addition, they are addressing issues on climate engineering and the IoT (internet of things), which can no longer be separated from IE and ethics of AI. Neuroethics is an emerging field. Combined with AI these techniques can enable developers – public or private – to abuse cognitive biases and trigger reactions and emotions without consent (Farisco, Salles, & Evers 2018:718). Whereas neural networks in AI are concerned with the development and application of algorithms for pattern recognition, classification, regression, and other task, neuroscience is a multidisciplinary fields of study that examines the ethical, legal, and social implications of neuroscience. Essentially, it encompasses the scientific study of the nervous system that regulates all human cognition, behaviour, and functions. The field encompasses a broad range of topics, from understanding how individual neurons function to exploring complex behaviours and cognitive processes. However, there is a link to AI because the more people discover how the nervous system works in relation to our world, the more there will be neuro-AI tangents (Berger & Rossi 2022:2054) together with the encompassing ethical challenges. An example of neuroscience and AI is wearable devices that integrate technologies from both neuroscience and AI to monitor, analyse, or interact with the brain and its functions. These wearables often serve various purposes, including research, healthcare, and personal wellbeing.

Based on their ability to collaborate across barriers such as borders, language, and cultural differences, organisations with the value and importance such as UNESCO, the OECD, and others are critical in researching and guiding society and the world in the ethical deployment of AI technologies. UNESCO emphasises the importance of promoting inclusivity in AI deployment for enhanced respect of cultural diversity. For this they provide ethical guidelines to address issues such as transparency, accountability, fairness, and the impact of AI on human rights. They advocate better education and capacity building specifically in the fields of IE and ethics of AI. However, their foremost value lies in the opportunities created to foster international cooperation and dialogue on AI ethics. This involves collaboration with other international organisations, governments, academia, industry, and civil society to develop shared principles and norms.

In the 1970s and 1980s, the OECD encouraged member countries to develop guidelines and legislation to manage data responsibly (Wright, De Hert, & Gutwirth 2011:119). In the seven-year stint between 1973 and 1980, one-third of the OECD’s 30 member countries enacted legislation intended to protect individuals against the abuse of data related to them and to give individuals the right of access to data with a view to checking their accuracy and appropriateness. The OECD is acutely aware of the tension that exists between the need of free flow of information and protecting of personal data (Wright et al. 2011:120). The 2013 privacy framework by the OECD and later in 2023 the OECD’s Good practice principles for data ethics in the public sector (OECD 2023), offer guardrails to the ethical use of information.

Pointers for Responsible AI

Table 4.3 below is based on an overview of literature, where specific deductions on essential elements that will ensure responsible AI use and development must be considered.

Table 4.3:Elements of responsible AI development and use

Following these guidelines helps to guarantee the responsible and ethical development and use of AI technology, with the aim of benefiting society while minimising any dangers and negative consequences.

Conclusion

To summarise, the field of AI ethics and its implementation in other areas like medicine and law, pose an intricate and dynamic challenge. The discussion about AI and the possibility of singularity, along with the implementation of fair information practices and responsible AI standards, highlight the need for a comprehensive approach to the development and regulation of AI.

We are currently at a critical point in time where the swift progress in AI technology, as demonstrated by systems such as GPT 4 and the expected development of models like GPT 5, necessitate a proactive and deliberate approach to ethical considerations. The OECD FIP standards serve as a fundamental framework for safeguarding data and ensuring privacy, which are of utmost importance in the era of AI. These principles, in conjunction with the guidelines for responsible AI, provide a clear path for the ethical, transparent, fair, and inclusive development of AI.

As we progress deeper into this era of technology, it is crucial to maintain a balance between innovation and accountability. AI development must be guided by ethical frameworks and norms to ensure that these technologies are utilised for the betterment of society while upholding human rights and dignity. Transparency, explainability, and accountability are not mere aspirations, but rather crucial prerequisites for fostering confidence and the approval of AI systems. To mitigate hazards connected with AI, it is necessary to address concerns of justice and nondiscrimination, prioritise privacy and data protection, and ensure safety and security.

Furthermore, the possible consequences of AI attaining or exceeding human cognitive capabilities – a central concept in the singularity discourse – emphasise the pressing need for these ethical deliberations. Continuous monitoring, stakeholder interaction, and adaptation is crucial in ensuring ethical practices in AI. The advancement of AI, with its potential benefits and difficulties, necessitate a cooperative endeavour involving technologists, ethicists, legislators, and the broader society. By engaging in such partnerships, we can guarantee that the development made in AI is in accordance with ethical values and has a good impact on the advancement of humanity.

Therefore, as we conclude this chapter, it is evident that the progression of AI ethics is continuous. The concepts and methods covered here are dynamic, continuously adapting to the technology they endeavour to regulate. The task at hand is not alone to responsibly advance AI, but also to consistently modify our ethical frameworks to align with the always evolving AI technology and its influence on society.

References

Alharbi, M. & Mukhari, A. 2023. Information science and interdisciplinary:

Literature review. International Journal of Advances in Science

Engineering and Technology 9(2):38-42.

Ahmed, SR. 2020. Preventing identity crime: Identity theft and identity

fraud: An identity crime model and legislative analysis with

recommendations for preventing identity crime. Leiden: Brill.

https://doi.org/10.1163/9789004395978

Asimov, I. (Ed.). 1984. Isaac Asimov presents the great SF stories 12 (1950). New York: DAW Books.

Bandi, A., Adapa, PVSR., & Kuchi, YEVPK. 2023. The power of generative AI: A review of requirements, models, input-output formats, evaluation metrics, and challenges. Future Internet 15, 260. 60 pages. https://doi.org/10.3390/fi15080260Bandi

Bartneck, C., Lütge, C., Wagner, A., & Welsh, S. 2020. An introduction to ethics in robotics and AI . Cham: Springer. https://doi. org/10.1007/978-3-030-51110-4

Bates, MJ. 2005. Information and knowledge: An evolutionary framework for information science. Information Research 10(4). 239. 30 pages.

Benk, M., Tolmeijer, S., Von Wangenheim, F., & Ferrario, A. 2022. The value of measuring trust in AI – a socio-technical system perspective. arXiv:2204.13480v1 . 12 pages. Available at: https:// arxiv.org/pdf/2204.13480.pdf . (Accessed on 23 January 2024).

Berger, SE. & Rossi, F. 2022. Addressing neuroethics issues in practice: Lessons learnt by tech companies in AI ethics. Neuron 110(13):2052 2056. https://doi.org/10.1016/j.neuron.2022.05.006

Bester, C. & Fischer, R. 2020. The essential relationship between information ethics and artificial intelligence. Artificial Intelligence, Ethics and Society 29:1-11. https://doi.org/10.29173/irie428

Birrer, FAJ. 1999. Understanding values and biases in IT. ACM SIGCAS Computers and Society 29(1):16-21. https://doi. org/10.1145/382042.382047

Borko, H. 1968. Information science: What is it? Journal of the Association for Information Science and Technology 19(1):3-5. https://doi. org/10.1002/asi.5090190103

Burgess, JTF. & Knox, EJM. (Eds.). 2019. Foundations of information ethics. Chicago: American Library Association.

Burrell, J. 2016. How the machine ‘thinks:’ Understanding opacity in machine learning algorithms. Big Data & Society 3(1). 12 pages. https://doi.org/10.1177/2053951715622512

Bynum, TW. 2010. Philosophy in the information age. Metaphilosophy 41(3):420-442. https://doi.org/10.1111/j.1467-9973.2010.01651.x

Capurro, R., Eldred, M., & Nagel, D. 2013. Digital whoness: Identity, privacy and freedom in the cyberworld . Frankfurt: Ontos Verlag. https:// doi.org/10.1515/9783110320428

Ess, C. 2008. Luciano Floridi’s philosophy of information and information ethics: Critical reflections and the state of the art. Ethics and Information Technology 10(2-3):89-96. https://doi.org/10.1007/ s10676-008-9172-8

Ess, C. 2014. Digital media ethics. Revised and updated 2nd ed. Cambridge: Polity.

Floridi, L. 1999. Information ethics: On the philosophical foundation of computer ethics. Ethics and Information Technology 1(1):33-52. https://doi.org/10.1023/A:1010018611096

Floridi. L. 2010. The Cambridge handbook of information and computer ethics. Cambridge: Cambridge University Press. https://doi.org/10.1017/ CBO9780511845239

Floridi, L. 2013. The philosophy of information. Oxford: Oxford University Press.

Floridi, L. 2014. The fourth revolution: How the infosphere is reshaping human reality. Oxford: Oxford University Press.

Floridi, L. 2015. The ethics of information. Oxford: Oxford University Press.

Floridi, L. 2016. Mature information societies – a matter of expectations. Philosophy & Technology 29(1):1-4. https://doi.org/10.1007/ s13347-016-0214-6

Floridi, L. 2018. Soft ethics and the governance of the digital. Philosophy & Technology 31(1):1-8. https://doi.org/10.1007/s13347-018-0303-9

Floridi, L. 2019. What the near future of artificial intelligence could be. Philosophy & Technology 32(1):1-15. https://doi.org/10.1007/ s13347-019-00345-y

Floridi, L. 2020. AI and its new winter: From myths to realities. Philosophy & Technology 33:1-3. https://doi.org/10.1007/s13347-020-00396 6

Floridi, L. 2024. On the future of content in the age of artificial intelligence: Some implications and directions. Philosophy & Technology 37. 112. 11 pages. https://doi.org/10.1007/s13347-024 00806-z

Farisco, M., Salles, A., & Evers, K. 2018. Neuroethics: A conceptual approach. Cambridge Quarterly of Healthcare Ethics 27(4):717-727. https://doi.org/10.1017/S0963180118000208

Friedman, B. & Nissenbaum, H. 1996. Bias in computer systems. ACM Transactions on Information Systems (Tois) 14(3):330-347. https:// doi.org/10.1145/230538.230561

Froehlich, TJ. 2000. Intellectual freedom, ethical deliberation and codes of ethics. IFLA Journal 26(4):264-272. https://doi. org/10.1177/034003520002600405

Han, J. 2022. An information ethics framework based on ICT platforms. Information 13. 440. 11 pages. https://doi.org/10.3390/ info13090440

Ibiricu, B. & Van der Made, ML. 2020. Ethics by design: A code of ethics for the digital age. Records Management Journal 30(3):395-414. https://doi.org/10.1108/RMJ-08-2019-0044

Mason, RO. 1986. Four ethical issues of the information age. MIS Quarterly 10(1):5-12. https://doi.org/10.2307/248873

Moor, JH. 1985. What is computer ethics? Metaphilosophy 16(4):266-275. https://doi.org/10.1111/j.1467-9973.1985.tb00173.x

Moor, JH. 2006. The nature, importance, and difficulty of machine ethics. IEEE Intelligent Systems 21:18-21. https://doi.org/10.1109/ MIS.2006.80

Moor, JH. 2020. What is computer ethics? In Miller, KW. & Taddeo, M. (Eds.): The ethics of information technologies, 15-24. London: Routledge. https://doi.org/10.4324/9781003075011-1

Morandín-Ahuerma, F. 2023. Twenty-three Asilomar principles for artificial intelligence and the future of life. OSF Preprints 5-27. https://doi.org/10.31219/osf.io/dgnq8 .

Ng, MY., Kapur, S., Blizinsky, KD., & Hernandez-Boussard, T. 2022. The AI life cycle: A holistic approach to creating ethical AI for health decisions. Nature Medicine 28(11):2247-2249. https://doi. org/10.1038/s41591-022-01993-y

Niederman, F., & Baker, EW. 2022. Ethics and AI issues: Old container with new wine? Information Systems Frontiers: A Journal of Research and Innovation 25(1):9-28. https://doi.org/10.1007/s10796-022 10305-1

OECD (Organisation for Economic Cooperation and Development). 2002. OECD guidelines on the protection of privacy and transborder flows of personal data. Available at: https://www.oecd.org/en/ publications/oecd-guidelines-on-the-protection-of-privacy and-transborder-flows-of-personal-data_9789264196391-en. html . (Accessed on 23 March 2024).

OECD (Organisation for Economic Cooperation and Development). 2013. The OECD privacy framework. Available at: https://www.oecd. org/sti/ieconomy/oecd_privacy_framework.pdf . (Accessed on 30 November 2023).

OECD (Organisation for Economic Cooperation and Development). 2023. Good practice principles for data ethics in the public sector. 10 August 2023. Available at: https://digital-skills-jobs.europa.eu/ en/inspiration/resources/oecds-good-practice-principles-data ethics-public-sector . (Accessed on 23 March 2024).

Paass, G. & Giesselbach, S. 2023. Foundation models for natural language processing: Pre-trained language models integrating media. Cham: Springer. https://doi.org/10.1007/978-3-031-23190-2

Quinn, MJ. 2011. Ethics for the information age. 4th ed. New York: Pearson.

Sartori, L. & Theodorou, A. 2022. A sociotechnical perspective for the future of AI: Narratives, inequalities, and human control. Ethics and Information Technology 24(4). 11 pages. https://doi. org/10.1007/s10676-022-09624-3

Shanahan, M. 2015. The technological singularity. Cambridge: MIT Press. https://doi.org/10.7551/mitpress/10058.001.0001

Shannon, CE. 1948. A mathematical theory of communication. The Bell System Technical Journal 27(3):379-423. https://doi. org/10.1002/j.1538-7305.1948.tb01338.x

Solove, DJ. 2006. A taxonomy of privacy. University of Pennsylvania Law Review 154(3):477-560. https://doi.org/10.2307/40041279

Solove, D. 2015. The growing problems with the sectoral approach to privacy law. Privacy + Security Blog . 13 November 2015. https:// teachprivacy.com/problems-sectoral-approach-privacy-law . (Accessed on 13 November 2024).

Stahl, BC. 2008. Discourses on information ethics: The claim to universality. Ethics and Information Technology 10:97-108. https://doi.org/10.1007/s10676-008-9171-9

Strickland, E. 2021. The turbulent past and uncertain future of AI: Is there a way out of AI’s boom-and-bust cycle? IEEE Spectrum 58(10):26-31. https://doi.org/10.1109/MSPEC.2021.9563956

Tavani, HT. 1999. Informational privacy, data mining, and the internet. Ethics and Information Technology 1(2):137-145. https://doi. org/10.1023/A:1010063528863

Tsamados, A., Aggarwal, N., Cowls, J., Morley, J., Roberts, H., Taddeo, M., & Floridi, L. 2021. The ethics of algorithms: Key problems and solutions. AI & Society 37:215-230. https://doi.org/10.1007/978 3-030-81907-1_8

Turing, AM. 1950. Computing machinery and intelligence. Mind 59:433 460. https://doi.org/10.1093/mind/LIX.236.433

Ulam, S. 1958. Tribute to John von Neumann. Bulletin of the American Mathematical Society 64(3):1-49. https://doi.org/10.1090/S0002 9904-1958-10189-5

Vaassen, B. 2021. AI, opacity, and personal autonomy. Philosophy & Technology 35. 88. 20 pages. https://doi.org/10.1007/s13347-022 00577-5

Wilson, TD. 1997. Information behaviour: An interdisciplinary perspective. Information Processing & Management 33(4):551-572. https://doi. org/10.1016/S0306-4573(97)00028-9

Weidinger, L., Mellor, J., Rauh, M., Griffin, C., Uesato, J., Huang, P-S., Cheng, M., Glaese, M., Balle, B., Kasirzadeh, A., Kenton, Z., Brown, S., Hawkins, W., Stepleton, T., Biles, C., Birhane, A., Haas, J., Rimell, L., Hendricks, LA., Isaac, W., Legassick, S., Irving, G., & Gabriel, I. 2021. Ethical and social risks of harm from language models. arXiv:2112.04359v1 . 64 pages. https://doi.org/10.48550/ arXiv.2112.04359

Woodward, B., Imboden, T., & Martin, NL. 2011. An undergraduate information security program: More than a curriculum. Journal of Information Systems Education 24(1):63-70.

Wright, D., De Hert, P., & Gutwirth, S. 2011. Are the OECD guidelines at 30 showing their age? Communications of the ACM 54(2):119-127. https://doi.org/10.1145/1897816.1897848

Young, J., Smith, TJ., & Zheng, SH. 2020. Call me BIG PAPA: An extension of Mason’s information ethics framework to big data. Journal of the Midwest Association for Information Systems (JMWAIS) 2:17-42.

Zhou, J., Chen, F., Berry, A., Reed, M., Zhang, S., & Savage, S. 2020. A survey on ethical principles of AI and implementations. IEEE Symposium Series on Computational Intelligence (SSCI), Canberra, ACT, Australia, 3010-3017. https://doi.org/10.1109/ SSCI47803.2020.9308437

Introduction

Goodbye to ChatGPT (chat generative pre-trained transformer), hello to AI (artificial intelligence) on the moon! AI is daring to have its finger touching the surface of the moon. The CMCSS (Canadian Mission Control Space Services) through budgetary funding of $3.04 million by the Canadian Space Agency made history when it launched the Rashid Rover on 11 December 2023, with the aim of spending one lunar day1 in space. The mission will see the Rover capturing and identifying geological features through pictures, and it was motivated by CMCSS’ urge to be the pioneer in showcasing AI’s DL (deep learning) capabilities first in lunar space. DL is a subset of ML (machine learning) and it relies on large and vast volumes of data, based on complex algorithms to train the model (Rane, Kaya, Mallick, & Rane 2024:218).

The enormous datasets, whether structured or unstructured, are where big data and PA (predictive analytics) traverse. Using data points and the precision of analysis are just a drop in an ocean in terms of capabilities and insights that can

One lunar day is equivalent to 29 Earth days.

be drawn from this interaction. The data gigantic is known as ‘big data,’ a term which, according to O’Leary (2013:96), has experienced exponential growth and has to date accumulated many definitions. Goh, Cheong, Tan, and Sharma (2023:16) allude that voluminous amounts of data that are being produced across various platforms, flourish exponentially. The authors go further to predict that this growth is expected to double every two years, and with such massive datasets, human-geared analysis is almost impractical, hence the need for sophisticated computational analysis (Emanuel & Wachter 2019:2281). To demonstrate the surge in data on the internet, Solarwinds Pingdom (2013) breaks down various types of datasets globally available on the web. For example, in 2012, there were approximately 2.2 billion e-mail users, 144 billion e-mail traffic per day, half a billion websites, 2.4 billion internet users, and over 100 million blogs. This constitutes data in unstructured form such as pictures, audio, text, and videos. As of 2023, there were over 5.3 billion internet users translating into almost 67% of the global population. In terms of the social media, there were about 4.95 billion users coupled with over 600 million blogs (Petrosyan 2023). These data statistics, according to Chu, Kim, Lin, Yu, Bradski, Ng, & Olukotun (2006:1 of 8) reflect data evolution through the ‘internet of people and things’ and the ‘internet of everything.’

The introduction of personal computers almost five decades ago (Flamm 1988:6), the prevalence of supercomputers and their interaction with human beings (Chen, Chen, & Lin 2020:75265), as well as the use of mobile technology (Alexandru, Alexandru, Coardos, & Tudora 2016:123) and social media have enabled the generation and availability of data in excessive forms (Luan, Geczy, Lai, Gobert, Yang, Ogata, Baltes, Guerra, Li, & Tsai 2020:2 of 11), which consequently paved way, and to some extent enabled these developments. According to Grand View Research (2023), the proliferating utilisation of mobile devices and their applications have transmogrified modern life in various ways, and this has caused a paradigm shift away from brick-and-mortar operations to digital and recently to cloud-based operations. Similarly, still on digital and technological transformation, Nam and Pardo (2011:185) highlight a number of evolutions in applications, ML techniques, DM (data mining), ICT (information and communication technology), as well as data fusion and pattern-recognition techniques as instrumental to the industrial revolution, with smart cities of the future being one part of it.

Big data as O’Leary (2013:97) puts it, were originally brought to the fore by Cox and Ellsworth in their 1997 work which reflected on how big data can be managed in the era of scientific visualisation (Cox & Ellsworth 1997). In that article, the two authors use the term ‘large data.’ Big data refers to information that is generated rapidly and in a wide variety from digital sources (Venugopal & Mamatha 2023:1). Conducting a literature review, an analysis of statistical data, mathematical, and statistical problem solving, improved personalised learning is purporting benefits driven by technological advancements in AI, big data, ML, and PA (Venugopal & Mamatha 2023:1). PA and AI rely on massive data warehouses that store both historical and real-time information. The connection between AI and big data can be best explained through a process of symbiosis, in that they are both dependent on one another to deliver above-par insights, where data analytic capabilities become an enabler in the journey for many industries to leverage the gains. For example, it is reported that the fusion of big data and ML tools and techniques such as DL models – CNNs (convolution neural networks) and LSTM (long-short term memory) architectures

– enable abilities of metropolitan municipal departments to

realise the emergence of smart cities, where they are able

to predict region-based traffic flow (Khan, Nazir, García-Magariño, & Hussain 2021). In these models CNNs are applied to classify spatial data whereas LSTM is used for temporal data classification (Khan et al. 2021:1-4 of 11). Other industrial areas benefiting from the application of AI, analytics, and big data include but are not limited to predictions of crowd flow (Zhou, Gu, Ling, Li, Zhuang, & Wang 2020:338), traffic flow (Khan, Ali, Ullah, & Bulbul 2018:71; Khan et al. 2021:1-2 of 11), water quality (Assem, Ghariba, Makrai, Johnston, Gill, & Pilla 2017:317), and air quality (Xiong, He, Huang, Yu, & Jing 2020:78). Neha and

Sidiq (2020:48) allude that within the educational landscape, IHEs (institutions of higher education) are able to predict

students’ performances and success through additional ML techniques such as CF (cooperative filtering) and recommender systems (RS). DM, regression techniques, classifications, neural networks, and SVMs (support vector machines) are some of the terminologies associated with ML and big data (Neha & Sidiq

2020:49).

No exploration of AI in education would be complete without addressing the challenges that arise in tandem with the integration of big data and PA. Algorithmic bias, ethical concerns, and the potential erosion of privacy are formidable issues that demand attention – hence they are discussed in the chapter. The chapter not only identifies these challenges, but quest also to highlight possible measures and principles aimed at mitigating against these challenges and concomitant risks that are complement to the implementation and integration of AI, big data, and PA education. Peering into the future, the concluding section of the chapter ventures into the implications that big data and AI clench for work and responsibilities of RERCs (research ethics review committees). The key questions worth exploring include the following:

These questions propel us towards an informed reflection on the evolving landscape of research ethics, underscoring the need for adaptable frameworks in the face of accelerating technological progress. This chapter is an exploration and exciting journey beaming through the past, present, and future of AI, big data, and PA integration in education. The threads that weave these technologies into the fabric and the nature of the educational sector are therefore unravelled, contemplating their impact on educators, research ethics review committees, and the very foundations of educational practices. The next section sets the scene for the chapter by delving into the classifications of AI systems.

Classifications of AI Systems

In this era of AI, where its capabilities and prowess are taking the world by storm, ML remains one of the topical issues within academic discussions and conferences (Sen, Hajra, & Ghosh 2020:99). The enhancement, efficiency, and precision of education practices can be attributed to how AI alongside data science and ML algorithms are leveraged. While data science, AI, and ML are closely connected to one another and can be clustered in the same domain, their overlap, specific applications, and meanings are worth being explored. Signifying the importance of ML techniques and their convergence with AI, Salem (2015:201) echoes sentiments in the literature that ML and computational intelligence coupled with DM techniques amplify the wherewithal of e-Iearning systems, thereby enabling them to reveal even more intelligent behaviour beneficial to the educational landscape.

The popularity of ML systems and techniques is credited to the work of Alan Turing, a British scientist who through his seminal work on computing, machines, and intelligence asked what could at the time be a rhetorical question: ‘Can a machine think?’ (Turing 1950). In his work, Turing hypothesises that it is possible that machines have the ability to think and execute decisions just as humans can. It goes without saying that he was mocked, however, little did they know that almost eight decades later, his thoughts would be poised to be the turning point for the AI of the future. Furthermore, his thoughts would continue to be the centre of debate between machines and human abilities. Muhammad and Yan (2015:946) point at the supernatural abilities of ML as computers are guided how to automatically explore good predictor drawing from previous experience. The authors go on to acclaim that abilities like that depend on how good the classifier is. Machines, referring to computers are extrapolated to be lacking abilities to learn from past experiences, while humans have that attribute (Muhammad & Yan 2015:947). This is an interesting reflection point to witness how far AI and in particular ML has evolved!

ML is built on the principle of incessant learning algorithms that continuously engage in decision-making or the identification

of patterns. ML models are trained to utilise preexisting data (this is one of the commonly used forms) or ML is a subset of AI empowered by neural network technology, and it is within the domain of data science. For example, the ChatGPT model by OpenAI is empowered by a transformer architecture that is used to generate text. This is the key strength and capability of generative AI (Gen-AI), and models like these are trained on enormous, large data, mainly testing the equivalent of over eight million documents in excess of 10 billion word characters (ATRIA 2023).

There are three main types of ML, namely supervised, unsupervised, and reinforcement learning. Big data, DM, and training are at the core of ML tools’ efficiency (Manne & Kantheti 2021:79), as all of these are using ML algorithms. Figure 5.1 below provides a schematic presentation of various classifications of AI systems. In comprehending the ML capabilities, Muhammad and Yan (2015:946) discern between supervised learning and unsupervised learning and explain that the former demonstrates the ability of machines to learn in a trained form. Supervised learning is one of the broad categories of ML, as it refers to algorithms and models that make predictions about future outcomes (Sen et al. 2020:100). These models draw their strengths and efficacy from the training they have been subjected to, which is based on previous data – hence they are referred to as ‘supervised’ models.

Unsupervised learning models on the other hand, are those models that are able to use machine intelligence to identify patterns and categories in the dataset. What sets them apart from supervised learning models is that they rely on unlabelled data and can independently organise that data to discover insights (Barlow 1989:295; Hastie, Tibshirani, & Friedman 2009:485-486).

Therefore, the distinguishing feature between supervised and unsupervised learning is the ability to act independently and is underpinned by data preparation and data preprocessing (Kotsiantis, Zaharakis, & Pintelas 2007:249). Lastly, reinforcement learning models are those models that act as intelligent agents and are aimed at taking actions and decisions within a well-defined environment. Each instance of datasets in ML is characterised by similar features, which could be continuous, categorical, or binary (Muhammad & Yan 2015:947). This is where ML and DL intersect and play an influential role. The two concepts are both characterised by a spectrum of approaches such as supervised, unsupervised, and reinforcement learning, neural networks, decision trees, and other forms of learning techniques (Samoili, López Cobo, Gómez, De Prato, Martinez-Plumed, & Delipetrev 2020:4).

NLP (natural language processing) applications and LLMs (large language models) are primarily conversational AI technologies, also known as dialogue or chatbot AI, focused on creating intelligent systems that can engage in natural language conversations with humans. These systems are designed to understand and generate humanlike responses to text or speech inputs. Conversational AI models employ techniques such as NLU (natural language understanding) to comprehend the user’s input, dialogue management to maintain context and flow within the conversation, and NLG (natural language generation) to generate appropriate and coherent responses. Their goal is to create interactive and engaging conversational experiences, whether in text-based chatbots, voice assistants (speech recognition), or other conversational interfaces. The other AI capabilities like expert systems, robotics, planning, and machine vision fall outside the scope of this chapter – however, they are worth exploring.

Advancements in Big Data, AI, and Predictive Analytics for Education

According to Cheng, Chen, Sun, Zhang, and Tao (2018), the developments and rise in numbers of technological tools propelled by AI has transformed various industries. At the heart of big data and AI capabilities lies the power of algorithms, which are also evident through the field of data science and PA. These sentiments are also shared by Goh et al. (2023:2) in a consultation paper prepared by a big data and AI review group on behalf of the bioethics advisory committee of Singapore. The authors put forward that developments in and the increasing use of big data, AT (auto tech) tools, and techniques enable researchers to merge and analyse large amounts of data, from which they are able to identify patterns, trends, and correlations. It is from these analyses that insights and data-driven decisions are drawn. Just as the other sectors or industries are reaping the benefits of AI and big data advancements, so is the education sector also witnessing a variety of benefits. The benefits for the education sector caused that IHEs are able to use PLA (predictive learning analytics) to track students’ learning progress, to enhance curriculum planning and development, to measure the efficacy of teaching strategies, and trigger data-driven decision-making and support for an enhanced teaching and learning environment (Zawacki-Richter, Marín, Bond, & Gouverneur 2019:3-4). Additionally, educators are efficient and effective in how they deliver content (Cheng et al. 2018).

The integration of AI, ML, and data analytics in education offers many benefits. Johnson (2023) adds that personalised learning strategies such as enabling early and tailored interventions in learning for struggling or at-risk students and promoting an adaptive learning environment to adjust course content to match the teaching styles are possible. The other benefits include stimulating a dynamic learning environment through which students’ answers can be assessed with the necessary effective teaching pedagogies (Lin, Huang, & Lu 2023). Naveenkumar (2023) highlights that the ability to adapt the learning process according to the needs and preferences of students makes it possible to provide personalised feedback. Goh et al. (2023:21) succinctly put some of these benefits as follows:

These advantages affirm the potential power of big data and AIML (artificial intelligence and machine learning). It is therefore valid that the why and how questions of big data and AI’s applicability in education are explored, with a detailed analysis being provided in the next four subsections.

Challenges and Ethical Implications of Using Big Data, AI, and Predictive Analytics Techniques

Big data and AI technologies offer an advantage of allowing users the ability to accumulate, process, and integrate vast amounts of data from a variety of sources in no time. This capability has yielded a number of benefits as discussed in the previous section. Through data analytics tools, researchers are now more than ever before able to access, integrate, and leverage expertise and knowledge across various disciplines – from physics, science, and computing, to mathematics, biology, and statistics. Within biomedicine and the health fraternity, medical diagnosis and clinical solutions are also pointing to the success of AI and big data in the field of health (Madani, Arnaout, Mofrad, & Arnaout 2018:1 of 8).

Despite the reported benefits, ethical dilemmas exist, and it is therefore important as Ben-Porath and Ben Shahar (2017:245) note that the implementing of big data and AI in education needs to be complemented by vigorous ethical frameworks and considerations to ensure a fair and transparent education process where equity is practised.

With ghost detection technologies, the downside here could be associated with false positives and false negatives. In the former, systems could fail to detect when the ghostwriter has indeed been utilised, whereas with the latter there could be false accusations of one purported to have used a ghostwriter when that was not the case. Lines (2016:889) warns that with an increasing and continuously evolving technology in the education space, digital plagiarism is on the rise, and subsequently ghostwriting services are proving and boost the potential of producing high quality services, thus making it difficult to detect. This is akin to essay mills and contract cheating as identified by Medway, Roper, and Gillooly (2018) in the work they did on covert investigation within the HE sector of the UK (United Kingdom). Sharing her findings, Lines is alarmed that the majority of these essay mills could have been easily submitted for assessment without any act of plagiarism being detected (Lines 2016:889). For the purpose of the current chapter, fairness and bias in algorithmic systems and decision-making, student privacy, and governance issues are ethical challenges that are in detail explored alongside their mitigating strategies in the next three sections.

The Future Implications of AI and Big Data on Research Ethics Review Committees

It is inevitable that industries are witnessing the increasing use of big data. However, just as this irresistible adoption is on the back of reported various benefits, there are equally associated implications. Luan et al. (2020:7 of 11) identify a number of major challenges and propose solutions in the area of research, policymaking, and industry. From a research perspective, the authors highlight that 1) the mode of teaching and learning is persistently transitioning to personalised and adaptive learning environments;

2) there is a single domain for educational AI research that is pro intelligent computing; and 3) machine-generated data need careful consideration redesign. For policy-making, the paradigm shift from traditional teaching methods and techniques to the digital world implies a shift in policies and guidelines, therefore there is a dire need for data privacy and the protection of personal information against unauthorised access and undue commercial exploitation. From an industrial perspective, commercialisation and high costs of intelligent tools remain a big challenge as these affect inclusivity and access.

Kanika and Khan (2017:195) cite security issues and warn that these issues may potentially invade one’s privacy – be it an individual or institution. Taking cognisance that these issues may well transcend into the HE landscape, it is important that a risk-oriented mindset and ethical lens in harnessing the potential of use of AI in the big data space is adopted. This position is premised on the potential impact of AI and big data on the work of RERCs.

The work of Ferretti, Ienca, Velarde, Hurst, and Vayena (2021b:138) sheds a light on challenges posed by big data research on the functions of RERCs, citing the magnitude of big data research projects. The authors argue that typical of these projects, they tend to have a scope that broadly requires a unique expertise, comprising new actors and making use of exceptional methodological approaches (Ferretti, Ienca, Sheehan, Blasimme, Dove, Farsides, Friesen, Kahn, Karlen, Kleist, & Liao 2021a:4 of 13). With this peculiarity, RERCs are typically constituted of skills and expertise from a variety of backgrounds (Nichols 2016:351), thus enabling the infusion of cross discipline (Garrard & Dawson 2005:419). Reviewers are grouped and get allocated content that match their strengths and expertise.

The aim of the ethical review process is to ensure that ethical guidelines during the research process are adhered to by researchers, to guard against unethical or illegal acts (Dove 2020:4). All in all, the end goal is to validate ethical procedures of the research being undertaken (Biagetti, Gedutis, & Ma 2020:1-2) to mitigate risks (Ferretti et al. 2021:8 of 13) and to reassure public and data subjects that the research process can be trusted (Resnik 2018:87).

In the era of AI and big data, it is important that the expertise of RERC members is assessed to ensure that review processes are not compromised. Once the process is compromised, trust and the validity of the research results and adopted methodologies are questioned. Ferretti et al. (2021a:8 of

13) identify persistent and novel weaknesses pertaining to failures of RERCs. Three weaknesses are identified, namely functional, purview, and persistent weaknesses, which are depicted in Figure 5.6. The functional and purview weaknesses are identified by the authors as novel, which implies that they are linked with traditional research ethics and are specific to the nature of big data projects (Ferretti et al. 2021a:3-4 of 13). The functional weaknesses stem from RERCs’ inabilities and inadequacies to review big data research projects. The purview weaknesses on the other hand relate to big data projects that may likely skew the horizon of RERCs.

of AI and big data on the work of RERCs. (Source:

Ferretti et al. 2021a:8 of 13)

The persistent weaknesses include a lack of transparency in the review process, capacity, a load to review projects, inconsistent review criteria, conflicting regulatory frameworks across jurisdictions, and a lack of an unsuitable environment to redistribute research risks and associated benefits. The proposed reforms to mitigate against these weaknesses are also indicated in the figure above, and these include regulatory, procedural, and complementary reforms (Ferretti et al. 2021a:8 of 13). Having discussed the implications brought by big data research and AI on the functions of RERCs, the conclusion focuses on the future trends that can be expected from the evolution of AI and related challenges.

Conclusion

The ascendancy and transformative power of big data, AI, and PA in the field of education has caused the ethics of research, teaching, and related procedures to be undoubtedly reshaped. The chapter delved in on the symbiotic relation between algorithms and education, exploring the nuances of their implementation.

These include writing style development detecting abilities, authorship verification, ghostwriter detection tools in academic writing, and AI-text generation tools. The chapter has dissected the intricate balance between algorithms and educational outcomes, which has shed a light on the implications and challenges associated with their integration into pedagogical practices. PA tools and their uses were discussed, including IESs, ITSs, and automated grading systems. From the perspective of research, responsive strategies on account of protecting and enhancing research and academic integrity were enacted. Through this lens, potential benefits, challenges, and possible governance measures of adopting and integrating big data, AI, and PA in education were discussed. These strategies are underpinned by the governance of AI, where protocols, guidelines and policy framework aimed at managing unethical conduct and malpractices in research were developed. This aligns with principles of responsible and ethical deployment, development, and application of AI tools. The work of RERCs was also not spared from this disruption. Various committees had to intensify issues relating to ethical consideration in line with their respective institutions’ research ethics policies and guidelines. That said, this necessitated a need to focus on operational informed consent in the era of AI, big data, and PA. Other related principles such as confidentiality, prevention of harm, data privacy, addressing bias and discrimination in algorithms, transparency, and accountability also had to be re-examined. This journey of reform meant that power dynamics in the field of ethics review processes and research governance has to adjust with the time. In essence, these advancements necessitated the need for stronger and robust ethical frameworks to empower RERCs to minimise and mitigate ethical blurred lines.

As far as challenges are concerned, ethical concerns dominate the discourse in the literature. For this chapter, I draw on weaknesses categorised as functional, purview, and persistent being documented by Ferretti et al. (2021a). The functional weaknesses relate to among others, a lack of proper expertise in regulating and reviewing research projects within the big data and PA space. These challenges are trailed by difficulties in operationalising the informed consent, following ex-ante review models and inability to comprehend unforeseeable risk assessments. On the purview category, challenges relate to difficulties of dealing with new actors conducting research, how to anonymise data and the broad scope of big data-related studies. The last category deals with persistency and challenges such as lack of transparency, conflicting regulations across various jurisdictions, and inconsistent review practices and criteria within and across various RERCs.

With these challenges, there are associated opportunities that arise. These may manifest themselves through reforms that may need to be instituted to mitigate risks and ethical concerns. These reforms can be categorised as regulatory, procedural, and complementary. The regulatory opportunities imply that there is a need to provide a new guidance on the practice of ethics in this new era. Furthermore, the purview of RERCs need to be expanded to bring new expertise. Given that AI transcends beyond borders, there is also a need to specify foundational global standards for the assessment of ethics. Global bodies like the OECD, European Commission on AI, and UNESCO (United Nations Educational, Scientific, and Cultural Organisation) have already started in this regard.

From a procedural perspective, there is a need to introduce new ethical assessment tools and systems to guide the behaviour of various role players such as researchers, ethics managers, and review members. This may also mean that the membership of the RERCs need re-examination and adjustment to ensure that there is a diversity of skills in the board, preferably the presence of external experts. From a complementary viewpoint, creating hybrid mechanisms for the review process will assist in mitigating the risks that come with the new era. This can be augmented by introducing new specialised committees to ensure that unfamiliar or too technical research projects related to AI and big data are properly assessed for ethical compliance. In parting shots, the multifaceted roles and capabilities of AI, big data, and PA in shaping the landscape have brought about challenges and opportunities. These changes mean that educators, research review committees, and educational institutions have to relook their governance processes, protocols, and policy frameworks.

References

Akgun, S. & Greenhow, C. 2022. Artificial intelligence in education: Addressing ethical challenges in K-12 settings. AI and Ethics 2(3):431-440. https://doi.org/10.1007/s43681-021-00096-7

Alexandru, A., Alexandru, CA., Coardos, D., & Tudora, E. 2016. Healthcare, big data and cloud computing. Management 1(2):123-131.

Ali, HIH. & Alhassan, A. 2021. Fighting contract cheating and ghostwriting in higher education: Moving towards a multidimensional approach. Cogent Education 8(1). 1885837. 18 pages. https://doi.or g/10.1080/2331186X.2021.1885837

Arrieta, AB., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., García, S., Gil-López, S., Molina, D., Benjamins, R., & Chatila, R., 2020. Explainable artificial intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information fusion 58:82-115. https://doi. org/10.1016/j.inffus.2019.12.012

Assem, H., Ghariba, S., Makrai, G., Johnston, P., Gill, L., & Pilla F. 2017. Urban water flow and water level prediction based on deep learning. In Altun, Y. (Ed.): Machine learning and knowledge discovery in databases , 317-329. Cham: Springer. https://doi. org/10.1007/978-3-319-71273-4_26

ATRIA. 2023. How does Chat GPT work? Atria Innovation. 5 January 2023. Available at: https://www.atriainnovation.com/en/how-does chat-gpt-work/ . (Accessed on 12 January 2024).

Baradwaj, BK. & Pal, S. 2012. Mining educational data to analyze students’ performance. International Journal of Advanced Computer Science and Applications , 2(6):63-69. http://dx.doi.org/10.14569/ IJACSA.2011.020609

Barlow, HB. 1989. Unsupervised learning. Neural Computation 1(3):295-311. https://doi.org/10.1162/neco.1989.1.3.295

Ben-Porath, S. & Ben Shahar, TH. 2017. Introduction: Big data and education: Ethical and moral challenges. Theory and Research in Education 15(3):243-248. https://doi. org/10.1177/1477878517737201

Bhushan, M., Shingate, RS., Vyas, T., Naman, T., & Shirsath, S. 2023. Intelligent tutoring system: Personalised learning plans with AI. Available at: https://www.researchgate.net/ publication/375696884_intelligent_tutoring_system_ personalised_learning_plans_with_ai . https://doi. org/10.13140/RG.2.2.36573.59369

Biagetti, MT., Gedutis, A., & Ma, L. 2020. Ethical theories in research evaluation: Exploratory. Scholarly Assessment Reports 2(1):1-9. https://doi.org/10.29024/sar.19

Biermann, OC. 2022. Writers want AI collaborators to respect their personal values and writing strategies: A human-centered perspective on AI co-writing. Doctoral thesis, University of British Columbia, Vancouver. Available at: https://open.library. ubc.ca/collections/ubctheses/24/items/1.0420422 . (Accessed on 15 June 2024).

bin Mat, U. & Buniyamin, N. 2017. Using neuro-fuzzy technique to classify and predict electrical engineering students’ achievement upon graduation based on mathematics competency. Indonesian Journal of Electrical Engineering and Computer Science 5(3):684 690. https://doi.org/10.11591/ijeecs.v5.i3.pp684-690

Bromley, J., Guyon, I., LeCun, Y., Säckinger, E., & Shah, R. 1993. Signature verification using a ‘siamese’ time delay neural network. Advances in Neural Information Processing Systems 6:737-744. https://doi.org/10.1142/S0218001493000339

Carbonell, JR. 1970. AI in CAI: An artificial-intelligence approach to computer-assisted instruction. IEEE Transactions on Man-Machine Systems 11(4):190-202. https://doi.org/10.1109/ TMMS.1970.299942

Chamola, V., Hassija, V., Sulthana, AR., Ghosh, D., Dhingra, D., & Sikdar,

B. 2023. A review of trustworthy and explainable artificial intelligence (XAI). IEEE Access 11:78994-79015. https://doi. org/10.1109/ACCESS.2023.3294569

Chassignol, M., Khoroshavin, A., Klimova, A., & Bilyatdinova, A. 2018. Artificial intelligence trends in education: A narrative overview. Procedia Computing Science 136:16-24. https://doi.org/10.1016/j. procs.2018.08.233

Chauhan, RK., Saharan, R., Singh, S., & Sharma, P. 2020. Automated content grading using machine learning. arXiv:2004.04300. 7 pages. https://doi.org/10.48550/arXiv.2004.04300

Chen, CP. & Zhang, CY. 2014. Data-intensive applications, challenges, techniques and technologies: A survey on big data. Information Sciences 275:314-347. https://doi.org/10.1016/j.ins.2014.01.015

Chen, L., Chen, P., & Lin, Z. 2020. Artificial intelligence in education: A review. IEEE Access 8:75264-75278. https://doi.org/10.1109/ ACCESS.2020.2988510

Cheng, Y., Chen, K., Sun, H., Zhang, Y., & Tao, F. 2018. Data and knowledge mining with big data towards smart production. Journal of Industrial Information Integration 9:1-13. https://doi.org/10.1016/j. jii.2017.08.001

Chicco, D. 2021. Siamese neural networks: An overview. In Cartwright,

H. (Ed.): Artificial neural networks, 73-94. Methods in Molecular Biology. Vol 2190. New York: Humana. https://doi. org/10.1007/978-1-0716-0826-5_3

Choi, Y. & McClenen, C. 2020. Development of adaptive formative assessment system using computerized adaptive testing and dynamic Bayesian networks. Applied Sciences 10(22):1-17. https:// doi.org/10.3390/app10228196

Chopra, S., Hadsell, R., & LeCun, Y. 2005. June. Learning a similarity metric discriminatively, with application to face verification. 2005 IEEE Computer Society Conference On Computer Vision And Pattern Recognition (CVPR’05) 1:539-546. https://doi.org/10.1109/ CVPR.2005.202

Chu, C-T., Kim, SK., Lin, Y-A., Yu, YY., Bradski, G., Ng, A. & Olukotun, K. 2006. Map-reduce for machine learning on multicore. Advances in Neural Information Processing Systems 19. 8 pages. Available at: http://books.nips.cc/nips19.html . (Accessed on 12 January 2024).

Cox, M. & Ellsworth, D. 1997. Managing big data for scientific visualization. IEEE, Proceedings of the 8th Conference on Visualization’97, 5-17.

Coyne, KP. 1986. Sustainable competitive advantage: What it is, what it isn’t. Business Horizons 29(1):54-61. https://doi. org/10.1016/0007-6813(86)90087-X

De Vries, K. 2022. Let the robot speak! AI-generated speech and freedom of expression. In Hindelang, S. & Moberg, A. (Eds.): YSEC Yearbook of Socio-Economic Constitutions 2021: Triangulating Freedom of Speech , 93-115. Cham: Springer. https://doi. org/10.1007/16495_2021_38

Deperlioglu, O. & Birtil, FS. 2016. Analysis of girls vocational high school students’ academic failure causes with data mining techniques. The Anthropologist 23(3):505-512. https://doi.org/10.1080/09720 073.2014.11891970

Dove, ES. 2020. Regulatory stewardship of health research: Navigating participant protection and research promotion. Cheltenham: Edward Elgar Publishing. https://doi.org/10.4337/9781788975353

Draxler, F., Werner, A., Lehmann, F., Hoppe, M., Schmidt, A., Buschek, D., & Welsch, R. 2023. The AI ghostwriter effect: Users do not perceive ownership of AI-generated text but self-declare as authors. arXiv preprint arXiv:2303.03283 . 40 pages. https://doi. org/10.1145/3637875

Eggers, WD., Schatsky, D., & Viechnicki, P. 2017. AI-augmented government: Using cognitive technologies to redesign public sector work. Deloitte Insights. 26 April 2017. Available at: https://www2.deloitte.com/us/en/insights/focus/cognitive technologies/artificial-intelligence-government.html . (Accessed on 21 January 2024).

Ekman, M. 2021. Learning deep learning: Theory and practice of neural networks, computer vision, natural language processing, and transformers using TensorFlow. Boston: Addison-Wesley Professional.

Emanuel, EJ. & Wachter, RM. 2019. Artificial intelligence in health care: Will the value match the hype? JAMA 321(23):2281-2282. https:// doi.org/10.1001/jama.2019.4914

Faggella, D. 2018. What is artificial intelligence? An informed definition. Emerj Insights . 21 December 2018. Available at: https://emerj. com/ai-glossary-terms/what-is-artificial-intelligence-an informed-definition/ . (Accessed on 13 January 2024).

Ferretti, A., Ienca, M., Sheehan, M., Blasimme, A., Dove, ES., Farsides, B., Friesen, P., Kahn, J., Karlen, W., Kleist, P., & Liao, SM. 2021a. Ethics review of big data research: What should stay and what should be reformed? BMC Medical Ethics 22(1). 51. 13 pages. https://doi.org/10.1186/s12910-021-00616-4

Ferretti, A., Ienca, M., Velarde, MR., Hurst, S., & Vayena, E. 2021b. The challenges of big data for research ethics committees: A qualitative Swiss study. Journal of Empirical Research on Human Research Ethics 17(1-2):129-143. https://doi/ org/10.1177/15562646211053538

Flamm, K. 1988. Creating the computer: Government, industry, and high technology. Washington DC: Brookings Institution.

Garrard, E. & Dawson, A. 2005. What is the role of the research ethics committee? Paternalism, inducements, and harm in research ethics. Journal of Medical Ethics 31(7):419-423. https://doi. org/10.1136/jme.2004.010447

Gehrmann, S., Dai, FZ., Elder, H., & Rush, AM. 2018. End-to-end content and plan selection for data-to-text generation. arXiv preprint arXiv:1810.04700 . 11 pages. https://doi.org/10.18653/v1/W18-6505

Goh, K-L., Cheong, VS., Tan, KCK., & Sharma, A. 2023. Ethical, legal and social issues arising from big data and artificial intelligence (AI) use in human biomedical research. New Castle University e-Prints. https://doi.org/10.57711/pgd8-ma17

Grand View Research. 2023. Digital transformation market to reach $4,617.78 billion by 2030. 8 August 2023. Available at: https:// www.grandviewresearch.com/press-release/global-digital transformation-market# . (Accessed on 7 December 2023).

Hastie, T., Tibshirani, R., & Friedman, J. 2009. Unsupervised learning. The elements of statistical learning: Data mining, inference, and prediction . New York: Springer. https://doi.org/10.1007/978-0 387-84858-7_14

Heikkilä, M. 2023. Why detecting AI-generated text is so difficult (and what to do about it). MIT Technology Review. 7 February 2023. Available at: https://www.technologyreview. com/2023/02/07/1067928/why-detecting-ai-generated-text is-so-difficult-and-what-to-do-about-it/ . (Accessed on 13 January 2024).

Hjuler, N. 2019. Algorithms and AI in education. Doctoral dissertation, University of Copenhagen, Faculty of Science, Computer Science Department, Copenhagen. Available at: https://researchprofiles. ku.dk/en/publications/algorithms-and-ai-in-education . (Accessed on 25 January 2024).

Hlosta, M., Herodotou, C., Papathoma, T., Gillespie, A., & Bergamin, P. 2022. Predictive learning analytics in online education: A deeper understanding through explaining algorithmic errors. Computers and Education: Artificial Intelligence 3. 100108. 12 pages. https:// doi.org/10.1016/j.caeai.2022.100108

Hung, BT. 2021. Face recognition using hybrid HOG-CNN approach. Research in Intelligent and Computing in Engineering: Select Proceedings of RICE 2020 , 715-723. https://doi.org/10.1007/978 981-15-7527-3_67

Janiesch, C., Zschech, P., & Heinrich, K. 2021. Machine learning and deep learning. Electronic Markets 31(3):685-695. https://doi. org/10.1007/s12525-021-00475-2

Jantawan, B. & Tsai, CF. 2013. The application of data mining to build classification model for predicting graduate employment. International Journal of Computer Science and Information Security 11(10):1-7. https://doi.org/10.arXiv:1312.7123

Johnson, R. 2023. The role of artificial intelligence in e-learning: Integrating AI tech into education. Atria Innovation. 27 October 2023. Available at: https://elearningindustry.com/role-of artificial-intelligence-in-elearning-integrating-ai-tech-into education . (Accessed on 14 January 2024).

Kanika, AA. & Khan, RA. 2017. Security integration in big data life cycle. In Singh, M., Gupta, P., Tyagi, V., Sharma, A., Ören, T., & Grosky, W. (Eds.): Advances in computing and data sciences, 192-200. ICACDS 2016. Communications in Computer and Information Science. Vol

721. Singapore: Springer. https://doi.org/10.1007/978-981-10 5427-3_21

Karabulut, EM., Özel, SA., & Ibrikci, T. 2012. A comparative study on the effect of feature selection on classification accuracy. Procedia Technology 1:323-327. https://doi.org/10.1016/j. protcy.2012.02.068

Khan, MA. & Vivek, MA. 2022. Artificial intelligence and big data:

The advent of new pedagogy in the adaptive e-learning system in the higher educational institutions of Saudi Arabia. Education Research International 2022:1-10. https://doi. org/10.1155/2022/1263555

Khan, S., Ali, H., Ullah, Z., & Bulbul, MF. 2018. An intelligent monitoring system of vehicles on highway traffic. 2018 12th International Conference on Open Source Systems and Technologies (ICOSST), 71 75. https://doi.org/10.1109/ICOSST.2018.8632192

Khan, S., Nazir, S., García-Magariño, I., & Hussain, A. 2021. Deep learning-based urban big data fusion in smart cities: Towards traffic monitoring and flow-preserving fusion. Computers & Electrical Engineering 89. 106906. 11 pages. https://doi.org/10.1016/j. compeleceng.2020.106906

Kirchner, JH., Ahmad, L., Aaronson, S., & Leike, J. 2023. New AI classifier for indicating AI-written text. OpenAI. 16 April 2023. Available at: https://openai.com/blog/new-ai-classifer-for-indicating-ai written-text . (Accessed on 4 January 2024).

Kotsiantis, SB., Zaharakis, I., & Pintelas, P. 2007. Supervised machine learning: A review of classification techniques. Informatica 31:249-268.

Kumar, AS. 2016. Edifice an educational framework using educational data mining and visual analytics. IJ Education and Management Engineering 2:24-30. https://doi.org/10.5815/ijeme.2016.02.03

Kumar, CK. 2018. Artificial Intelligence: Definition, types, examples, technologies. 31 August 2018. Available at : https:// chethankumargn.medium.com/artificial-intelligence definition-types-examples-technologies-962ea75c7b9b . (Accessed on 9 January 2024).

Leslie, D. 2019. Understanding artificial intelligence ethics and safety: A guide for the responsible design and implementation of AI systems in the public sector. Zenoda . 11 June 2019. https://doi. org/10.5281/zenodo.3240529

Liddy, ED. 2001. Natural language processing. Encyclopaedia of Library and Information Science. 2nd ed. New York: Marcel Decker.

Lin, CC., Huang, AYQ., & Lu, OHT. 2023. Artificial intelligence in intelligent tutoring systems toward sustainable education: A systematic review. Smart Learning Environments 10(1). 22 pages. https://doi. org/10.1186/s40561-023-00260-y

Lin, CC., Kuo, CH., & Chiang, HT. 2021. CNN-based classification for point cloud object with bearing angle image. IEEE Sensors journal 22(1):1003-1011. https://doi.org/10.1109/JSEN.2021.3130268

Lin, CJ. & Mubarok, H. 2021. Learning analytics for investigating the mind map-guided AI chatbot approach in an EFL flipped speaking classroom. Educational Technology and Society 24(4):16-35.

Lines, L. 2016. Ghostwriters guaranteeing grades? The quality of online ghost-writing services available to tertiary students in Australia. Teaching in Higher Education 21(8):889-914. https://doi.org/10.10 80/13562517.2016.1198759

Lorenzen, SS., Hjuler, NOD., & Alstrup, S. 2019. Investigating writing style development in high school. Proceedings of the 12th International Conference on Educational Data Mining (EDM 2019), 572-575. https://doi.org/10.48550/arXiv.1906.03072

Luan, H., Geczy, P., Lai, H., Gobert, J., Yang, SJ., Ogata, H., Baltes, J., Guerra, R., Li, P., & Tsai, CC. 2020. Challenges and future directions of big data and artificial intelligence in education. Frontiers in Psychology 11. 580820. 11 pages. https://doi. org/10.3389/fpsyg.2020.580820

Ma, Y. & Guo, G. (Eds.). 2014. Support vector machines applications. Cham: Springer Science. https://doi.org/10.1007/978-3-319-02300-7

Madani, A., Arnaout, R., Mofrad, M., & Arnaout, R. 2018. Fast and accurate view classification of echocardiograms using deep learning. NPJ Digital Medicine 1(1). 8 pages. https://doi.org/10.1038/s41746 017-0013-1

Manne, R. & Kantheti, SC. 2021. Application of artificial intelligence in healthcare: Chances and challenges. Current Journal of Applied Science and Technology 40(6):78-89. https://doi.org/10.9734/ cjast/2021/v40i631320

Martinho-Truswell, E. 2018. How AI could help the public sector. Harvard Business Review . 26 January 2018. Available at: https://hbr. org/2018/01/how-ai-could-help-the-public-sector . (Accessed on 14 January 2024).

Medway, D., Roper, S., & Gillooly, L. 2018. Contract cheating in UK

higher education: A covert investigation of essay mills. British Educational Research Journal 44:393-418. https://doi.org/10.1002/ berj.3335

Mehr, H. 2017. Artificial intelligence for citizen services and government. Available at: https://ash.harvard.edu/files/ash/files/artificial_ intelligence_for_citizen_services.pdf . (Accessed on 16 January 2024).

Mishra, R. 2019. Usage of data analytics and artificial intelligence in ensuring quality assurance at higher education institutions. 2019 Amity International Conference on Artificial Intelligence (AICAI), 1022-1025. https://doi:10.1109/AICAI.2019.8701392

Mishra, T., Kumar, D., & Gupta, S. 2016. Students’ employability prediction model through data mining. International journal of applied engineering research 11(4):2275-2282.

Moltzau, A. 2020. Estonia’s national strategy for artificial-intelligence. 2 January 2020. Available at: https://medium.com/swlh/estonias national-strategy-for-artificial-intelligence-2623259ddf4c . (Accessed on 10 January 2024).

Muhammad, I. & Yan, Z. 2015. Supervised machine learning approaches: A survey. ICTACT Journal on Soft Computing 5(3):946-952. https:// doi.org/10.21917/ijsc.2015.0133

Murty, MN. & Raghava, R. 2016. Support vector machines and perceptrons: Learning, optimization, classification, and application to social networks . Cham: Springer. https://doi.org/10.1007/978-3-319 41063-0

Nam, T. & Pardo, TA. 2011. Smart city as urban innovation: Focusing on management, policy, and context. Proceedings of the 5th International Conference on Theory and Practice of Electronic Governance , 185-194. https://doi.org/10.1145/2072069.2072100

Naveenkumar, S. 2023. Transforming education with machine learning: Personalized learning, intelligent tutoring, and educational games. 25 February 2023. Available at: https://www.linkedin. com/pulse/transforming-education-machine-learning personalized-intelligent-s/ . (Accessed on 10 January 2024).

Neha, K. & Sidiq, SJ. 2020. Analysis of student academic performance through expert systems. International Research Journal on Advanced Science Hub 2(Special Issue ICIES 9S):48-54. https:// doi.org/10.47392/irjash.2020.158

Nichols, AS. 2016. Research ethics committees (RECS)/institutional review boards (IRBS) and the globalisation of clinical research: Can ethical oversight of human subjects research be standardized? Washington University Global Studies Law Review 15:351-379.

O’Leary, DE. 2013. Artificial intelligence and big data. IEEE intelligent systems 28(2):96-99. https://doi.org/10.1109/MIS.2013.39

OECD (Organisation for Economic Co-operation and Development). 2023. The impact of AI on the workplace: Main findings from the OECD AI surveys of employers and workers. Available at: https://www. oecd.org/employment-outlook/2023/#ai-jobs. (Accessed on 12 December 2023).

OECD (Organisation for Economic Co-operation and Development). 2024. Principles for trustworthy AI. Available at: https://oecd.ai/en/ai principles . (Accessed on 12 December 2023).

Papamitsiou, Z. & Economides, AA. 2014. Learning analytics and educational data mining in practice: A systematic literature review of empirical evidence. Journal of Educational Technology & Society 17(4):49-64. https://doi.org/10.1111/bjet.12747

Patgiri, R. & Ahmed, A. 2016. Big data: The V’s of the game changer paradigm. 2016 IEEE 18th International Conference on High-Performance Computing and Communications, IEEE 14th International Conference on Smart City, and IEEE 2nd International Conference on Data Science and Systems (HPCC/SmartCity/DSS), Sydney, 17-24. https://doi:10.1109/HPCC-SmartCity-DSS.2016.0014

Perkins, M. 2023. Academic integrity considerations of AI large language models in the post-pandemic era: ChatGPT and beyond. Journal of University Teaching & Learning Practice 20(2). 24 pages. https:// doi.org/10.53761/1.20.02.07

Perset, K. 2024. A pioneering year for AI and the OECD. OECD-AI. 9 January 2024. Available at: https://oecd.ai/en/wonk/2023. (Accessed on 14 January 2024).

Petrosyan, A. 2023. Internet and social media users in the world 2023. Statista . 25 October 2023. Available at: https://www.statista.com/ statistics/617136/digital-population-worldwide/ . (Accessed on 12 January 2024).

Poola, I. 2023. Overcoming ChatGPT’s inaccuracies with pre-trained AI prompt engineering sequencing process. International Journal of Technology and Emerging Sciences 3(3):16-19.

Qian, C., He, T., & Zhang, R. 2017. Deep learning based authorship identification. Available at: https://www.semanticscholar.org/ paper/DeepLearning-based-Authorship-Identification-QianHe/ ab0ebe094ec0a44fb0013d640b344d8cfd7adc81 . (Accessed on 15 January 2024).

Quan, Z. & Pu, L. 2023. An improved accurate classification method for online education resources based on support vector machine (SVM): Algorithm and experiment. Education and Information Technologies 28(7):8097-8111. https://doi.org/10.1007/s10639 022-11514-6

Rane, J., Kaya, O., Mallick, SK., & Rane, NL. 2024. Smart farming using artificial intelligence, machine learning, deep learning, and ChatGPT: Applications, opportunities, challenges, and future directions. Generative Artificial Intelligence in Agriculture, Education, and Business , 218-272. https://doi.org/10.70593/978 81-981271- 7-4_6

Remaida, A., Moumen, A., El Bouzekri El Idrissi, Y., & Abdellaoui, B. 2021. Tuning convolutional neural networks hyperparameters for offline handwriting recognition. Proceedings of the 2nd International Conference on Big Data, Modelling and Machine Learning (BML 2021) , 71-76. https://doi. org/10.5220/0010728600003101

Remian, D. 2019. Augmenting education: Ethical considerations for incorporating artificial intelligence in education. Master’s dissertation, faculty of the Instructional Design Master’s Degree Program University of Massachusetts at Boston. Available at: https://scholarworks.umb.edu/instruction_capstone/52. (Accessed on 15 January 2024).

Resnik, DB. 2018. The ethics of research with human subjects: Protecting people, advancing science, promoting trust. Cham: Springer. https://doi/org/10.1007/978-3-319-68756-8

Salem, ABM. 2015. Towards of intelligence education and learning. IEEE 7th International Conference on Intelligent Computing and Information Systems (ICICIS), Cairo , 196-202. https://doi.org/10.1109/ IntelCIS.2015.7397221

Samad, AA., Arshad, MM., & Siraj, MM. 2021. Towards enhancement of privacy-preserving data mining model for predicting students learning outcomes performance. IEEE International Conference on Computing (ICOCO), November 2021, 13-18. https://doi. org.10.1109/ICOCO53166.2021.9673544

Samoili, S., López Cobo, M., Gómez, E., De Prato, G., Martinez-Plumed, F., & Delipetrev, B. 2020. AI watch – defining artificial intelligence

– towards an operational definition and taxonomy of artificial intelligence. Luxembourg: Publications Office of the European Union. https://data.europa.eu/doi/10.2760/382730

Sen, PC., Hajra, M., & Ghosh, M. 2020. Supervised classification algorithms in machine learning: A survey and review. In Mandal,

J. & Bhattacharya, D. (Eds.): Emerging technology in modelling and graphics: Advances in intelligent systems and computing, 99-111. Singapore: Springer. https://doi.org/10.1007/978-981-13-7403 6_11

Sharma, A. & Mansotra, V. 2019. Deep learning based student emotion recognition from facial expressions in classrooms. International Journal of Engineering and Advanced Technology 8(6):4691-4699. https://doi.org/10.35940/ijeat.F9170.088619

Sheth, A. 2014. Transforming big data into smart data: Deriving value via harnessing volume, variety, and velocity using semantic techniques and technologies. IEEE 30th International Conference on Data Engineering (ICDE), Chicago, 2014 , 2. https://doi:10.1109/ ICDE.2014.6816634

Sivarajah, U., Kamal, MM., Irani, Z., & Weerakkody, V. 2017. Critical analysis of big data challenges and analytical methods. Journal of Business Research 70:263-286. https://doi.org/10.1016/j. jbusres.2016.08.001

Solarwinds Pingdom. 2012. Internet 2011 in numbers. 17 January 2012. Available at: http://royal.pingdom.com/2012/01/17/internet2011 in-numbers . (Accessed on 12 January 2024).

Solarwinds Pingdom. 2013. Internet 2012 in numbers. 17 January 2013. Available at: https://www.pingdom.com/blog/internet-2012-in numbers . (Accessed on 12 January 2024).

Stamatatos, E. 2009. A survey of modern authorship attribution methods. Journal of the American Society for information Science and Technology 60(3):538-556. https://doi.org/10.1002/asi.21001

Stavngaard, M., Sørensen, A., Lorenzen, S., Hjuler, N., & Alstrup, S. 2019. Detecting ghostwriters in high schools. arXiv preprint arXiv:1906.01635. 6 pages.

Taigman, Y., Yang, M., Ranzato, MA., & Wolf, L. 2014. Deepface: Closing the gap to human-level performance in face verification. Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), 1701-1708. https://doi.org/10.1109/ CVPR.2014.220

Takyar, A. 2022. AI use cases & applications across major industries. LeewayHertz-Software Development Company Publications. Available at: https://www.leewayhertz.com/ai-use-cases-and applications/ . (Accessed on 12 December 2023).

Tjano, RN. 2021. An empirical study of corporate governance and sustainability reporting practices in South African state-owned entities. Doctoral thesis. University of South Africa. Pretoria. Available at: https://uir.unisa.ac.za/handle/10500/28099. (Accessed on 12 January 2024).

Turing, AM. 1950. Can a machine think. Mind 59(236):433-460. https:// doi.org/10.1093/mind/LIX.236.433

Venugopal, R. & Mamatha, V. 2023. Impact of artificial intelligence (AI) on teaching and learning in India’s higher education sector. IOSR Journal of Research & Method in Education 13(5):1-6. https://doi. org/10.9790/7388-1305020106

Wahle, JP., Ruas, T., Kirstein, F., & Gipp, B. 2022. How large language models are transforming machine-paraphrased plagiarism. Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP) , 1-12. https://doi.org/10. ArXiv:2210.03568

Wang, F. & Preininger, A. 2019. AI in health: State of the art, challenges, and future directions. Yearbook of Medical Informatics 28(1):16-26. https://doi.org/10.1055/s-0039-1677908

Wirtz, BW., Weyerer, JC., & Geyer, C. 2019. Artificial intelligence and the public sector – applications and challenges. International Journal of Public Administration 42(7):596-615. https://doi.org/10.1080/0 1900692.2018.1498103

Xiong, Y., He, Y., Huang, H., Yu, C., & Jing, X. 2020. Air quality statistics and prediction based on urban agglomerations and sentiment analysis of people under different pollutants. In Wang, Y., Fu, M., Xu, L., & Zou, J. (Eds.): Signal and information processing, networking and computers. lecture notes in electrical engineering, 78-87. Vol 628. Singapore: Springer. https://doi. org/10.1007/978-981-15-4163-6_10

Zawacki-Richter, O., Marín, VI., Bond, M., & Gouverneur, F. 2019. Systematic review of research on artificial intelligence applications in higher education – where are the educators? International Journal of Educational Technology in Higher Education 16(39):1-27. https://doi.org/10.1186/s41239-019-0171-0

Zhang, C., Jiang, W., Zhang, Y., Wang, W., Zhao, Q., & Wang, C. 2022. Transformer and CNN hybrid deep neural network for semantic segmentation of very-high-resolution remote sensing imagery. IEEE Transactions on Geoscience and Remote Sensing 60:1-20. https://doi.org/10.1109/TGRS.2022.3144894

Zhou, Q., Gu, J-J., Ling, C., Li, W-B., Zhuang, Y., & Wang, J. 2020. Exploiting multiple correlations among urban regions for crowd flow prediction. Journal of Computer Science and Technology 35:338-352. https://doi.org/10.1007/s11390-020-9970-y

Zhu, Z., Lin, K., Jain, AK., & Zhou, J. 2023. Transfer learning in deep reinforcement learning: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence 45(11):13344-13362. https://doi. org/10.1109/TPAMI.2023.3292075

Zimmerman, AA. 2023. Ghostwriter for the masses: ChatGPT and the future of writing. Annals of Surgical Oncology 30:3170-3173. https://doi.org/10.1245/s10434-023-13436-0

Introduction

Educational practices in the 21st century have undergone a significant transformation due to technological advances, particularly in artificial intelligence (AI). Recent developments and expansion in machine learning led to the emergence of generative artificial intelligence (Gen-AI), an advanced and innovative technology (Hu 2022). Lim, Gunasekara, Pallant, Pallant, and Pechenkina (2023:2 of 13) define Gen-AI as a technology that

1) uses deep learning models to 2) generate human-like content (e.g., images and words) in response to 3) complex and varied prompts (e.g., languages, instructions, and questions). It can do so in response to a wide range of prompts that may be complex and diverse, including different languages, instructions, or questions. In other words, Gen-AI can produce human-like outputs when given various types of input.

ChatGPT (chat generative pre-trained transformer), an implementation of Gen-AI, developed and released by OpenAI in November 2022, uses publicly available digital content data to produce text in multiple languages that is very similar to human writing. OpenAI released ChatGPT 3.5 in November, followed by a more powerful and capable version to perform more complex functions, ChatGPT 4, in March 2023. In January 2024, it had around 180 million users and is a rapidly growing and widely used platform (OpenAI 2024). Upon registration, both versions are readily available for use without any prior training required. While the first version is free, ChatGPT 4 currently costs $20 monthly (OpenAI 2024). The powerful language model, which is still an emerging innovation, has already shown that it can display creativity in its writing, from a single paragraph to a complete research paper, on almost any topic (Aydin & Karaarslan 2022:22). It has made educational resources more accessible and extended the reach of higher education beyond traditional classroom settings or local settings, creating new opportunities for both lecturers and students (Celik 2023:4).

The scalability and accessibility of Gen-AI enable a global reach in HE (higher education), breaking geographical barriers and making quality education available to students worldwide. These developments further represent a significant advancement in smart learning, enhancing the ability to offer personalised and engaging learning experiences (Pesovski, Santos, Henriques, & Trajkovik 2024:2). ChatGPT plays a pivotal role in enhancing the learning environment by tailoring responses to each student’s unique needs and preferences, as highlighted in a recent study by Rawas (2023:2). Furthermore, the capabilities of ChatGPT, such as automated grading and feedback, ease the burden on educators, enabling them to focus on more valuable tasks, such as facilitating discussions, building social engagements, and providing specialised support that cannot be provided by Gen-AI (Conijn, Kahr, & Snijders 2023:41).

Despite the capabilities of ChatGPT (and similar technologies such as Google’s Bard and Copilot by Microsoft), it should be considered along with the risks and challenges to ensure its ethical and responsible use in HE. For example, Adiguzel, Kaya, and Cansu (2023:1) mention ethical issues such as bias and the need to train educators to successfully incorporate Gen-AI tools as challenges in using ChatGPT. Additionally, Rawas (2023:2) adds that due to the significant amount of data ChatGPT uses, privacy and security issues are a top priority when using the tool. As ChatGPT is an AI-generated assistant that facilitates communication between students and educators, it is important to ensure that its use does not have negative consequences. To achieve this, adequate monitoring by educators and accountability on the side of students are needed. This will ensure that all interactions are appropriate, respectful, ethical, and safe. Therefore, it is the responsibility of both students and educators to use Gen-AI tools such as ChatGPT with care and consideration to avoid any intentional and unintentional harm that may arise from using the tool. In doing so, the tool remains useful to all users.

The use of Gen-AI tools such as ChatGPT in HE has gained significant attention from universities worldwide as a disruptive tool for teaching and learning, as well as for supporting their staff and students (e.g., Conijn et al. 2023:39; Wong 2024:2). Similarly, IHEs (institution of higher education) have actively explored the potential of incorporating Gen-AI tools into their pedagogical approaches. The aim is to enhance the quality of education by offering personalised learning experiences that address the unique needs of each student (Rasul, Nair, Kalendra, Robin, De Oliveira Santini, Ladeira, Sun, Day, Rather, & Heathcote 2023:43). Such an approach recognises that Gen-AI technology has the potential to transform traditional teaching methods, making learning more interactive, engaging, and effective. With the help of Gen-AI tools such as ChatGPT, IHEs can create customised learning paths that are tailored to the individual needs, interests, and learning styles of each individual student.

By analysing student performance data and feedback, ChatGPT algorithms can identify strengths and areas that need more support. This will, in turn, provide them with targeted interventions to help them overcome their challenges. Such personalised learning approaches can help students stay motivated and engaged, which could lead to better outcomes. However, due to the recent emergence of Gen-AI tools, the academic community is still in the process of determining the most efficient and responsible ways to incorporate such tools into HE. Ongoing research in this area is needed.

This chapter offers insights into how Gen-AI tools like ChatGPT could revolutionise HE. From a theoretical and interpretative analysis based on an integrative literature review (Hambaloyi & Jordan 2016:121), recent articles (mostly published in 2023) are focusing on the transformative role of Gen-AI, with specific reference to how ChatGPT was analysed in HE. The following search string was used on Google Scholar to search for articles: Higher education AND generative AI OR ChatGPT AND opportunities AND transformation AND teaching AND learning. From the literature search, the limited investigation into the topic was confirmed, as noted by Rawas (2023:3). Most studies revolve around broader AI in education, education in general, or the use of specific tools. To further refine the search, articles focussing on these topics were excluded for the purpose of this chapter, although they might have had relevance. Based on the literature review, this chapter aims to investigate the unique opportunities and challenges that ChatGPT presents for transforming HE.

Opportunities of ChatGPT in Higher Education

The emergence of Gen-AI tools is the key to progress innovation and can revolutionise HE. It can create exciting new opportunities for students, educators, and IHEs alike if they understand its possibilities and realities. With its cutting-edge technology and advanced natural language processing capabilities, ChatGPT can open previously unexplored avenues of teaching and learning.

Based on the literature reviewed (as indicated below), this section discusses some of the promising options that ChatGPT provides. The studies below meet the inclusion criteria and provide an insightful glimpse into how ChatGPT is changing, and could increasingly revolutionise teaching and learning in HE in the future. However, it is worth mentioning that these studies are not meant to be an exhaustive list of all research done in 2023 in this area. The table below shows the authors and their respective countries for the selected papers. Of importance is that, although the search string was limited to opportunities, the chosen articles also highlight challenges that come along with them to provide a balanced perspective. The opportunities and challenges identified in the articles are listed below, and a discussion follows.

Lastly, in addition to the work of the above researchers, the following question was asked to ChatGPT 4: Considering its opportunities and challenges, how could ChatGPT transform teaching and learning in higher education? The response included the opportunities and challenges indicated in Table 6.2.

Table 6.2:Opportunities and challenges identified by ChatGPT

Based on the concepts provided in Tables 6.1 and 6.2, ChatGPT created the following word cloud on the opportunities of ChatGPT. It indicated that it used the Python application to create it. The word cloud visually represents the key terms and concepts, with more frequent terms appearing larger in the cloud. It provides a quick and insightful view of the main themes and opportunities associated with ChatGPT.

Similar to the opportunities, ChatGPT was asked to create a word cloud for the challenges listed in the table.

The purpose of the word clouds was to extract and visualise key information from the table presenting ChatGPT’s benefits and challenges in transforming HE. The most prominent or recurring concepts are clearly visible as they appear in the larger text. The word clouds also helped to identify the most important aspects of the opportunities in the following discussion.

Conclusion

This chapter aimed to offer insights in how Gen-AI tools such as ChatGPT could transform HE. It used an integrative literature review and focused on opportunities and challenges. The findings revealed that the transformative potential of ChatGPT in HE is significant and will continue to play a crucial role in shaping HE. From the literature review, several opportunities were discussed, namely personalised learning and support, improved student engagement and motivation, individualised feedback, assistance with writing and research, automated administrative support, innovative evaluation, educational content creation, language learning support, and expansion of access. With these opportunities, various challenges were also identified. These are ethical issues and bias, privacy and security concerns, transparency and accountability, a lack of human interaction and empathy, an overreliance on technology, academic integrity, technical and connectivity issues, and inaccurate information.

AI tools such as ChatGPT have already begun transforming HE and will increasingly change how educators teach and students learn. However, it is only the beginning, and several areas still need to be addressed. As AI tools will only get better and more human-like, HE will have to be adapted. Although ChatGPT has the potential to improve teaching and learning for everyone, the key is finding a balance and using this tool to enhance, rather than replace traditional educational practices. Collaboration among educators, researchers, policymakers, and students is essential to ensure that AI is used ethically and responsibly in education. The goal is to create a more equitable and effective education system that provides students with personalised teaching, feedback, and support by addressing the challenges posed by AI technologies and leveraging their benefits.

Every study has limitations, and this study is no exception. The main limitations are related to the number of articles reviewed and the methodology used. However, the findings of research papers from different countries around the world provided meaningful insights into the topic that can assist educators, students, and IHEs in a better understanding of Gen-AI in HE.

Lastly, ChatGPT and similar technologies will have an increasing impact on HE in the future. Therefore, further empirical research is needed to determine the potential benefits of these emerging technologies and how they can be used to the benefit of all.

References

Adiguzel, T., Kaya, MH., & Cansu, FK. 2023. Revolutionizing education with AI: Exploring the transformative potential of ChatGPT. Contemporary Educational Technology 15(3):1-13. https://doi. org/10.30935/cedtech/13152

Aydın, Ö. & Karaarslan, E. 2022. OpenAI ChatGPT generated literature review: Digital twin in healthcare. In Aydın, Ö. (Ed.): Emerging computer technologies Vol 2, 22-31. İzmir: İzmir Akademi Dernegi. https://dx.doi.org/10.2139/ssrn.4308687

Celik, I. 2023. Towards Intelligent-TPACK: An empirical study on teachers’ professional knowledge to ethically integrate artificial intelligence (AI)-based tools into education. Computers in Human Behavior 138:1-12. https://doi.org/10.1016/j.chb.2022.107468

Chan, CKY. 2023. A comprehensive AI policy education framework for university teaching and learning. International Journal of Educational Technology in Higher Education 20(38):1-25. https:// doi.org/10.1186/s41239-023-00408-3

Chiu, TK., Moorhouse, BL., Chai, CS., & Ismailov, M. 2023. Teacher support and student motivation to learn with artificial intelligence (AI) based chatbot. Interactive Learning Environments 1-17. https://doi. org/10.1080/10494820.2023.2172044

Conijn, R., Kahr, P., & Snijders, C. 2023. The effects of explanations in automated essay scoring systems on student trust and motivation. Journal of Learning Analytics 10(1):37-53. https://doi. org/10.18608/jla.2023.7801

Gill, SS., Xu, M., Patros, P., Wu, H., Kaur, R., Kaur, K., Fuller, S., Singh, M., Arora, P., Parlikad, AK., & Stankovski, V. 2024. Transformative effects of ChatGPT on modern education: Emerging era of AI chatbots. Internet of Things and Cyber-Physical Systems 4:19-23. https://doi.org/10.1016/j.iotcps.2023.06.002

Hambaloyi, W. & Jordan, P. 2016. Systematic review as a research method

in postgraduate nursing education. Health SA Gesondheid 21(0):120-128. http://doi.org/10.1016/j.hsag.2015.08.002

Hu, L. 2022. Generative AI and future. Medium. 17 July 2022. Available at: https://pub.towardsai.net/generative-ai-and-future c3b1695876f2 . (Accessed on 4 January 2024).

Hutson, J. & Schnellmann, A. 2023. The poetry of prompts: The collaborative role of generative artificial intelligence in the creation of poetry and the anxiety of machine influence. Global Journal of Computer Science and Technology: 23(1):1-14. https:// digitalcommons.lindenwood.edu/faculty-research-papers/462/

Kerr, P. 2016. Adaptive learning. ELT Journal 70(1):88-93. https://doi. org/10.1093/elt/ccv055

Lim, WM., Gunasekara, A., Pallant, JL., Pallant, JI., & Pechenkina, E. 2023. Generative AI and the future of education: Ragnarök or reformation? A paradoxical perspective from management educators. The International Journal of Management Education 21. 100790. 13 pages. https://doi.org/10.1016/j.ijme.2023.100790

Lo, CK., Hew, KF., & Jong, MSY. 2024. The influence of ChatGPT on student engagement: A systematic review and future research agenda. Computers & Education 219. 105100. 21 pages. https://doi. org/10.1016/j.compedu.2024.105100

Mhlanga, D. 2023. Digital transformation education, opportunities, and challenges of the application of ChatGPT to emerging economies. Education Research International 2023(4):1-13. https://doi. org/10.1155/2023/7605075

OpenAI. 2024. ChatGPT (Large language model). Available at: https://chat. openai.com . (Accessed on 15 January 2024).

Pesovski, I., Santos, R., Henriques, R., & Trajkovik, V. 2024. Generative AI for customizable learning experiences. Sustainability 16(3034):1 23. https://doi.org/10.3390/su16073034

Raschka, S. 2023. What are the different approaches for detecting content generated by LLMs such as ChatGPT? And how do they work and differ? Blog. Available at: https://sebastianraschka.com/ blog/2023/detect-ai.html . (Accessed on 14 January 2024).

Rasul, T., Nair, S., Kalendra, D., Robin, M., De Oliveira Santini, F., Ladeira, WJ., Sun, M., Day, I., Rather, RA., & Heathcote, L. 2023. The role of ChatGPT in higher education: Benefits, challenges, and future research directions. Journal of Applied Learning and Teaching 6(1):41-56. https://doi.org/10.37074/jalt.2023.6.1.29

Rawas, S. 2023. ChatGPT: Empowering lifelong learning in the digital age of higher education. Education and Information Technologies 29(6):1-14. https://doi.org/10.1007/s10639-023-12114-8

Rudolph, J., Tan, S., & Tan, S. 2023. ChatGPT: Bullshit spewer or the end of traditional assessments in higher education? Journal of Applied Learning and Teaching 6(1):342-363. https://doi. org/10.37074/jalt.2023.6.1.9

Sok, S. & Heng, K. 2024. Opportunities, challenges, and strategies for using ChatGPT in higher education: A literature review. Journal of Digital Educational Technology 4(1):1-11. https://doi.org/10.30935/ jdet/14027

Sullivan, M., Kelly, A., & McLaughlan, P. 2023. ChatGPT in higher education: Considerations for academic integrity and student learning. Journal of Applied Learning and Teaching 6(1):31-40. https://doi.org/10.37074/jalt.2023.6.1.17

Tiwari, CK., Bhat, MA., Khan, ST., Subramaniam, R., & Khan, MAI. 2024. What drives students toward ChatGPT? An investigation of the factors influencing the adoption and usage of ChatGPT. Interactive Technology and Smart Education 21(3):333-355. https:// doi.org/10.1108/ITSE-04-2023-0061

Vargas-Murillo, AR., De la Asuncion, INM., & De Jesús Guevara-Soto, F. 2023. Challenges and opportunities of AI-assisted learning: A systematic literature review on the impact of ChatGPT usage in higher education. International Journal of Learning, Teaching and Educational Research 22(7):122-135. https://doi.org/10.26803/ ijlter.22.7.7

Vogels, EA. 2021. Some digital divides persist between rural, urban and suburban America. Pew Research Center. Available at: https:// www.pewresearch.org/fact-tank/2021/08/19/some-digital divides-persist-between-rural-urban-and-suburban-america/ . (Accessed on 12 January 2024).

Wong, WKO. 2024. The sudden disruptive rise of generative artificial intelligence? An evaluation of their impact on higher education and the global workplace. Journal of Open Innovation: Technology, Market, and Complexity 10(2):1-8. https://doi.org/10.1016/j. joitmc.2024.100278

Yang, TC., Hwang, GJ., & Yang, SJH. 2013. Development of an adaptive learning system with multiple perspectives based on students’ learning styles and cognitive styles. Journal of Educational Technology & Society 16(4):185-200.

Introduction

AI (Artificial intelligence) in education offers a wide range of benefits that can enhance the learning experience for students, streamline administrative processes, and support educators. AI refers to advanced computing systems that process the ability to perform tasks that traditionally require human intelligence. These systems can learn, adapt, correct, synthesis, and use large amounts of text to carry out complex tasks (Popenici & Kerr 2017:2). AI has opened new possibilities and encouraged challenges at all levels of education (Silander & Stigmar 2019:274). Its applications in education are on the rise and have recently received much attention. However, AI is not new and has a long history. The origin of AI can be traced to the 1950s, when John McCarthy arranged a two-month workshop at Dartmouth College in the USA. In the proposal for the workshop, McCarthy was the first person to use the term ‘Artificial Intelligence’ in 1956 (Kühl, Schemmer, Goutier, & Satzger 2022:2237). Gen-AI (Generative AI) refers to technology that uses DL (deep learning) models to generate human-like content, like images and text, based on complex and varied prompts, including languages, instructions, and questions (Lim, Gunasekara, Pallant, Pallant, & Pechenkina 2023:2 of 13). Currently, Gen-AI is at the forefront of leveraging DL models to mimic human-like content.

Since the release of Gen-AI models such as ChatGPT (chat generative pre-trained transformer), Copilot, and Meta AI in 2022, much has been written about their promises and challenges in education. However, most of the research involved literature reviews (e.g., Bozkurt 2023; Grassini 2023; Baidoo-Anu & Owusu Ansah 2023), while not much empirical research on how educators regard and use these tools has been published. For this reason, this chapter is based on empirical research by investigating the world of Gen-AI as seen through the lens of a specific group of educators. The chapter explores and reflects on Gen-AI in education from the perspectives of those at its heart – the educators. It not only presents their perspectives on the diverse benefits and challenges of Gen-AI but emphasises the necessity for a balanced, ethical, and informed approach to harnessing Gen-AI’s capabilities. The aim is to pave the way for Gen-AI to complement educators and, in so doing, equally benefit educators and students.

Literature Review

Methodology

Findings and Discussion

Conclusion

The purpose of this chapter was to report on students’ perceptions of the benefits and challenges of Gen-AI tools in their respective educational contexts. Based on an interpretative paradigm, we used an exploratory case study design to analyse the responses to questions on the discussion forum of a master’s module in education. The study provides valuable insights from educators who actively engage with Gen-AI tools. The most prominent benefits identified were that these digital tools can provide individualised, 24/7 guidance and support and lesson content. They also shared that these tools can assist with simplifying language, a unique benefit that could not be found in previous studies. Furthermore, Gen-AI tools provide opportunities for active involvement. They save time and make information and knowledge accessible. The participants’ main challenges were that the technologies could provide biased information, lead to decreased human interaction and lack EI. Additionally, it could provide inaccurate information and has a risk that students can over-rely on these tools. They further identified the risks of plagiarism and data privacy and security as challenges. Lastly, some participants provided guidelines on how these technologies could be approached and how the risks could be mitigated. These guidelines focused on the need for policies and AI literacy training, setting an example and providing guidelines to students in the use of Gen-AI. An additional important point raised was that Gen-AI tools should complement, and not replace, the human elements of teaching and learning, ensuring a wholistic and inclusive educational approach.

While the study provides important insights, it comes with limitations that must be considered. The main limitation of this study is the small number of participants and specific context in which they were studied. It is possible that if a different sample had been used, the findings could have been different. Additionally, the study relied on a single instrument to collect data, which may not have provided a comprehensive understanding of the responses. As a result, more empirical research with the same and different groups and different instruments could lead to a deeper understanding of how Gen-AI tools can be used to benefit all its users equally, ultimately leading to improved learning outcomes.

References

Ai, Y., Hu, Y., & Zhao, W. 2024. Interactive learning: Harnessing technology to alleviate foreign language anxiety in foreign literature education. Interactive Learning Environments 2023:1-20. https://doi.org/10.1080/10494820.2024.2409990

Akgun, S. & Greenhow, C. 2022. Artificial intelligence in education: Addressing ethical challenges in K-12 settings. AI and Ethics 2(3):431-440. https://doi.org/10.1007/s43681-021-00096-7

Anderson, O. 2024. Ethical considerations in machine translation: Bias, fairness, and accountability. Innovative Computer Sciences Journal 10(1):1-7. https://doi.org/10.48047/resmil.v10i1.22

Baidoo-Anu, D. & Owusu Ansah, L. 2023. Education in the era of generative artificial intelligence (AI): Understanding the potential benefits of ChatGPT in promoting teaching and learning. Journal of AI 7(1):52-62. https://doi.org/10.61969/jai.1337500

Bozkurt, A. 2023. Unleashing the potential of generative AI, conversational agents and chatbots in educational praxis: A systematic review and bibliometric analysis of GenAI in education. Open Praxis 15(4):261-270. https://doi.org/10.55982/openpraxis.15.4.609

Braun, V. & Clarke, V. 2012. Thematic analysis. In Cooper, H., Camic, PM., Long, DL., Panter, AT., Rindskopf, D., & Sher, KJ. (Eds.): APA handbook of research methods in psychology. Vol. 2: Research designs: Quantitative, qualitative, neuropsychological, and biological, 57-71. 1st ed. Washington DC: American Psychological Association.

Cassidy, C. 2023. Australian universities to return to ‘pen and paper’ exams after students caught using AI to write essays. The Guardian . 10 January 2023. Available at: https://www. theguardian.com/australia-news/2023/jan/10/universities-to return-to-pen-and-paper-exams-after-students-caught using-ai-to-write-essays . (Accessed on 12 January 2024).

Chen, P., Wu, L., & Wang, L. 2023. AI fairness in data management and analytics: A review on challenges, methodologies and applications. Applied Sciences 13(18):1-33. https://doi.org/10.3390/ app131810258

Cheuk, T. 2021. Can AI be racist? Color-evasiveness in the application of machine learning to science assessments. Science Education 105(5):825-836. https://doi.org/10.1002/sce.21671

Dai, CP. & Ke, F. 2022. Educational applications of artificial intelligence in simulation-based learning: A systematic mapping review. Computers and Education: Artificial Intelligence 3:1-17. https://doi. org/10.1016/j.caeai.2022.100087

De Vos, AS., Delport, CSL., Fouché, C., & Strydom, H. 2011. Research at grass roots: A primer for the social science and human professions. Pretoria: Van Schaik.

Garvey, C. & Maskal, C. 2020. Sentiment analysis of the news media on artificial intelligence does not support claims of negative bias against artificial intelligence. Omics: A Journal of Integrative Biology 24(5):286-299. http://doi.org/10.1089/omi.2019.0078

Grassini, S. 2023. Shaping the future of education: Exploring the potential and consequences of AI and ChatGPT in educational settings. Education Sciences 13. 692. 13 pages. https://doi.org/10.3390/ educsci13070692

Karpouzis, K., Pantazatos, D., Taouki, J., & Meli, K. 2024. Tailoring education with GenAI: A new horizon in lesson planning. arXiv preprint: 2403.12071. 10 pages. https://doi.org/10.48550/ arXiv.2403.12071

Kivunja, C. & Kuyini, AB. 2017. Understanding and applying research paradigms in educational contexts. International Journal of Higher Education 6(5):26-41. https://doi.org/10.5430/ijhe.v6n5p26

Kumar, H. & Martin, A. 2023. Artificial emotional intelligence: Conventional and deep learning approach. Expert Systems with Applications 212. 118651. 20 pages. https://doi.org/10.1016/j. eswa.2022.118651

Kühl, N., Schemmer, M., Goutier, M., & Satzger, G. 2022. Artificial intelligence and machine learning. Electronic Markets 32(4):22352244.

Kung, TH., Cheatham, M., Medenilla, A., Sillos, C., De Leon, L., Elepaño, C., Madriaga, M., Aggabao, R., Diaz-Candido, G., Maningo, J., & Tseng, V. 2023. Performance of ChatGPT on USMLE: Potential for AI-assisted medical education using large language models. PLoS Digital Health 2(2). 198. 12 pages. https://doi.org/10.1371/journal. pdig.0000198

Lietz, CA., Langer, CL., & Furman, R. 2006. Establishing trustworthiness in qualitative research in social work: Implications from a study regarding spirituality. Qualitative Social Work 5(4):441-458. https://doi.org/10.1177/1473325006070288

Lim, WM., Gunasekara, A., Pallant, JL., Pallant, JI., & Pechenkina, E. 2023. Generative AI and the future of education: Ragnarök or reformation? A paradoxical perspective from management educators. The International Journal of Management Education 21(2). 13 pages. https://doi.org/10.1016/j.ijme.2023.100790

Lincoln, YS. & Guba, EG. 1985. Naturalistic inquiry. Newbury Park: Sage. https://doi.org/10.1016/0147-1767(85)90062-8

Marouf, A., Al-Dahdooh, R., Ghali, MJA., Mahdi, AO., Abunasser, BS., & Abu-Naser, SS. 2024. Enhancing education with artificial intelligence: The role of intelligent tutoring systems. International Journal of Engineering and Information Systems 8(8):10-16.

McCombs, M. & Valenzuela, S. 2020. Setting the agenda: Mass media and public opinion. Hoboken: John Wiley.

OpenAI. 2024. ChatGPT (large language model). Available at: https://chat. openai.com . (Accessed on 5 January 2024).

Popenici, SAD. & Kerr, S. 2017. Exploring the impact of artificial intelligence on teaching and learning in higher education. Research and Practice in Technology Enhanced Learning 1:3-13. https://doi.org/10.1186/s41039-017-0062-8

Rasul, T., Nair, S., Kalendra, D., Robin, M., De Oliveira Santini, F., Ladeira, WJ., Sun, M., Day, I., Rather, RA., & Heathcote, L. 2023. The role of ChatGPT in higher education: Benefits, challenges, and future research directions. Journal of Applied Learning and Teaching 6(1):33-57. https://doi.org/10.37074/jalt.2023.6.1.29

Ray, PP. 2023. ChatGPT: A comprehensive review on background, applications, key challenges, bias, ethics, limitations and future scope. Internet of Things and Cyber-Physical Systems 3:121-154. https://doi.org/10.1016/j.iotcps.2023.04.003

Rudolph, J., Tan, S., & Tan, S. 2023. ChatGPT: Bullshit spewer or the end of traditional assessments in higher education? Journal of Applied Learning and Teaching 6(1):342-362. https://doi.org/10.37074/ jalt.2023.6.1.9

Saunders, S. 2023. Rather than ban generative AI, universities must learn from the past. University World News. 24 February 2023. Available at: https://www.universityworldnews.com/post. php?story=20230221075136410 . (Accessed on 12 January 2024).

Silander, C. & Stigmar, M. 2019. Individual growth or institutional development? Ideological perspectives on motives behind Swedish higher education teacher training. Higher Education 77:265-281. https://doi.org/10.1007/s10734-018-0272-z

Sullivan, M., Kelly, A., & McLaughlan, P. 2023. ChatGPT in higher education: Considerations for Academic Integrity and Student Learning. Journal of Applied Learning and Teaching 6(1):31-40. https://doi.org/10.37074/jalt.2023.6.1.17

Tlili, A., Shehata, B., Adarkwah, MA., Bozkurt, A., Hickey, DT., Huang, R., & Agyemang, B. 2023. What if the devil is my guardian angel: ChatGPT as a case study of using chatbots in education. Smart Learning Environments 10(1). 24 pages. https://slejournal. springeropen.com/articles/10.1186/s40561-023-00237-x

Van den Berg, G. & Du Plessis, E. 2023. ChatGPT and generative AI: Possibilities for Its contribution to lesson planning, critical thinking and openness in teacher education. Educational Sciences 13:1-12. https://doi.org/10.3390/educsci13100998

Yin, RK. 2009. Case study research: Design and methods. Vol. 5. London: Sage.

Introduction

Online assessment techniques have become increasingly popular recently, especially with the rise of COVID-19 (Coronavirus disease of 2019), since they offer practical and effective ways to access knowledge, skills, and competencies. 2023 will go down in history as the year in which AI (artificial intelligence) – more significantly, LLMs (large language models) – and their brazen poster child, ChatGPT (chat generative pre-trained transformer), have revolutionised everything (Bekker 2024). The introduction of advanced LLMs like ChatGPT, developed by OpenAI, can potentially transform online assessment significantly. However, while there are ongoing discussions about ChatGPT, more practical examples need to be provided to showcase its utilisation in online assessments. It is crucial to shed light on the assessment process, as many students focus solely on passing exams rather than embracing the learning experience throughout the year, leading to a disconnect between the process and the outcome (MacGregor 2023).

Thanks to the swift growth of AI, the current field of education has access to brand-new, cutting-edge solutions. ChatGPT, a LLM that can create text that resembles human language and participate in interactive conversations, is one such technology. Incorporating ChatGPT and other related LLMs replaces intriguing opportunities and challenges as online assessment becomes more common. Combining AI and LLMs like ChatGPT introduces a revolution in online assessment. AI-powered applications enable automated grading and individualised feedback and improve efficiency (Celik, Dindar, Muukkonen, & Järvelä 2022:617). However, integrating AI-powered tools raises concerns and challenges about the fairness, validity, and originality of material accessed in online learning environments (Hoyos 2023). For instance, ChatGPT is built on OpenAI ChatGPT 3, which can create narratives, develop computer codes, translate and complete calculations, and autofill images (Rudolph, Tan, & Tan 2023:342). This chapter examines the potential effects of adopting ChatGPT in online assessment by considering its benefits and potential drawbacks for educators and institutions. The chapter also suggests potential answers to address these problems and propose strategies for online assessment.

Conceptual Framework

Understanding ChatGPT and Online Assessment

GPT is a LLM utilised in various NLP (natural language processing) applications, including ChatGPT (Dheda 2023). GPT plays a vital role in ChatGPT by enabling the chatbot to generate responses that closely resemble human language when users interact with it. ChatGPT is a specific variation of GPT 3, an LLM developed by OpenAI (Cotton, Cotton, & Shipway 2023:1 of 12). Since its introduction in 2021 and November 2022, ChatGPT has attracted much interest from the tech and media sectors. The transformer design is the foundation for GPT 3, which has become widely used in various NLP tasks (Cotton, Cotton, & Shipway 2023:1 of 12). With 175 billion parameters, GPT 3 stands out for its massive size, making it one of the most essential LLMs on the market today (Gpt Generative Pretrained Transformer, Thunström, & Steingrimsson 2022:3 of 7; Sümer 2023:2). It is well known for its ability to perform a wide range of language-related tasks, including translation, summarisation, question answering, and text generation, frequently with little to no specialised training needed for each task.

Since its introduction, GPT 3 has been used in several fields, such as language translation, content creation, and language modelling. It is noteworthy that GPT 3 has proven to be highly accurate at translating across languages and producing cohesive and educational summaries of significant publications (Brown, Mann, Ryder, Subbiah, Kaplan, Dhariwal, Neelakantan, Shyam, Sastry, Askell, Agarwal, Herbert-Voss, Krueger, Henighan, Child, Ramesh, Ziegler, Wu, Winter, Hesse, Chen, Sigler, Litwin, Gray, Chess, Clark, Berner, McCandlish, Radford, Sutskever, & Amodei 2020:7). Furthermore, GPT 3’s ability to comprehend and react to natural language inputs has been demonstrated by the development of chatbots that can talk with people and answer enquiries. The excellent abilities of GPT 3 have drawn much interest and inspired discussions on the prospective applications and effects of AI in language processing (Cotton et al. 2023:1 of 12).

Hsuan-Tien Lin, a professor in the Department of Computer Science and Information Engineering at NTU (the National Taiwan University), argues that when introducing ML (machine learning) to information retrieval or other fields, ‘we need to be careful in understanding what the technique is about to ensure the best use’ (Sharma 2024). Instead, many researchers rush to use ML without comprehensively understanding its techniques and limitations. They may not think clearly about training, validation, and test sets and ‘end up with a rough instead of a rigorous use of machine learning methods’ (Chih-Jen Lin, quoted in Sharma 2024). Sharma (2024) notes that ‘[t]he phenomenon of rough use of machine learning methods is common and sometimes unavoidable. AI moves quickly from research laboratories to real-world applications, yet ML models that work well in the laboratory can fail in real-world use, with essential consequences.’

It is crucial to understand ChatGPT’s capabilities and constraints to gauge its possible influence on online assessment now that it is built on GPT 3.5 or GPT 4. Section 2 will give an overview of ChatGPT, highlighting its ability to produce contextually appropriate responses and its potential to participate in interactive conversations. It also presents the idea of online assessment, which has grown in acceptance due to its practicality and scalability, underscoring the requirement for efficient and flexible assessment techniques.

Discussion and Strategies on the Use of ChatGPT in Online Assessment

An essential component of HE is assessment, used to gauge students’ learning and development. A vast array of learning outcomes, including knowledge, abilities, and attitudes can be evaluated using a variety of assessment formats, such as exams, papers, projects, and presentations. Developing customised tests is one possible use for GPT 3 in HE. According to each student’s unique requirements and skills, GPT 3 could be used to create tailored tests or quizzes (Barber et al. 2021:24; Zawacki-Richter, Marín, Bond, & Gouverneur 2019:4). GPT 3 could be used to develop questions that are specific to each student’s level of proficiency and that challenge them to exhibit their knowledge and skills, which could be especially helpful in courses that emphasise language skills or critical thinking (Bommasani, Hudson, Adeli, Altman, Arora, Von Arx, Bernstein, Bohg, Bosselut, Brunskill, Brynjolfsson , Buch, Card, Castellon, Chatterji, Chen, Creel, Davis, Demszky, Donahue, Doumbouya, Durmus, Ermon, Etchemendy, Ethayarajh, Fei-Fei, Finn, Gale, Gillespie , Goel, Goodman, Grossman, Guha, Hashimoto, Henderson, Hewitt, Ho, Hong, Hsu, Huang, Icard, Jain, Jurafsky, Kalluri, Karamcheti, Keeling, Khani, Khattab, Koh, Krass, Krishna, Kuditipudi, Kumar, Ladhak, Lee, Lee, Leskovec, Levent, Li, Li, Ma, Malik, Manning, Mirchandani, Mitchell, Munyikwa, Nair, Narayan, Narayanan, Newman, Nie, Niebles, Nilforoshan, Nyarko, Ogut, Orr, Papadimitriou, Park, Piech, Portelance, Potts, Raghunathan, Reich, Ren, Rong, Roohani, Ruiz, Ryan, Ré, Sadigh, Sagawa, Santhanam, Shih, Srinivasan, Tamkin, Taori, Thomas, Tramèr, Wang, Wang, Wu, Wu, Wu, Xie, Yasunaga, You, Zaharia, Zhang, Zhang, Zhang, Zhang, Zheng, Zhou, & Liang 2021:29).

Developing interactive, game-based tests is another possible use for GPT 3 in HE. Chatbots or virtual assistants that push students to solve challenges or respond to enquiries using natural language interaction could be developed using GPT 3. In addition to teaching students useful communication and problem-solving skills, this might be an entertaining and exciting opportunity for them to demonstrate their knowledge and abilities. Students can receive personalised answers to their enquiries and instant feedback from chatbot applications. These chatbots may also personalise the educational process by suggesting books and websites based on students’ interests and requirements. To help students better understand the topic, chatbot applications can also offer educational resources like study guides and lecture notes (Perez, Massey-Allard, Butler, Ives, Bonn, Yee, & Roll 2017:290; Cotton et al. 2023:2, 7 of 12).

A more effective and individualised learning experience could be made possible by using ChatGPT to assess assignments and give students feedback in real-time (Gao 2021:322-324; Cotton et al. 2023:3, 5 of 12; Zawacki-Richter et al. 2019:4). GPT 3, for instance, might be used to grade essays and other written work, freeing up educators to concentrate on higher-level duties like giving students feedback and support. Additionally, GPT 3 could be used to grade tests or quizzes more rapidly and precisely, giving students feedback in real-time (Cotton et al. 2023:2 of 12).

Due to the difficulties in assessing student submissions generated by ChatGPT, there are various strategies that educators can employ to address these issues. As a starting point, instructors can offer comprehensive and explicit guidelines to students organising their assignments (Barile 2023). They might also utilise straightforward and specific language, reiterate instructions detailing the objective of the task, and include examples. This approach can help to ensure that assignments are composed in a more organised and logical fashion. Additionally, educators can implement a rubric to assess the quality of students’ work. Rubrics establish a clear set of evaluation criteria for measuring performance across various activities, including assignments, presentations, and overall productivity. Rubrics typically include criteria, performance levels, and descriptors (Leadership Workshop 2012). This approach aids in accurately evaluating a student’s effort and comprehension of the subject matter. Additionally, educators are able to apply automated and manual assessment methods to gauge the student’s grasp of the material (Cotton et al. 2023:5 of 12). This ensures that the true extent of the student’s understanding is effectively assessed.

Additionally, academic personnel should search for linguistic abnormalities or patterns: Because chatbots frequently have low language skills, they may produce more human-like content, contain repetitious words or phrases, or use strange or inconsistent terminology. One way to determine whether a chatbot created a piece of writing is to look at its language (Sümer 2023).

Chatbots cannot generate novel concepts or conduct original research. Consequently, it is doubtful that material produced by a chatbot will contain accurate citations or references to sources (Gewirtz 2023). One way to determine if a chatbot wrote a piece of work is to look at its sources and citations. Educators therefore ought to verify originality. Chatbots cannot produce original work. As a result, anything produced by a chatbot is probably going to be rather similar to previously published sources. It is possible to determine whether a chatbot most likely created a piece of writing by using plagiarism detection software or manual examination to check for originality (Gewirtz 2023).

Although LLMs can produce writing that makes sense, they might only sometimes do so. Examining the essay for factual mistakes or discrepancies may reveal that it was written by a machine (Cotton et al. 2023:5 of 12). The educator should also check grammar and spelling: AI-generated writing may be more error-free than human writing, which may contain typos and grammatical faults. However, the quality of the LLM and the input data used to train it can affect this (Sümer 2023).

It is advisable to employ language analysis tools because some of them (like the GPT 2 output detector demo) are made to examine written work and identify patterns or abnormalities that could suggest that a chatbot created the work (Cotton et al. 2023:4 of 12; Sümer 2023).

When creating scenarios for real-world case study-based tests, academic users may also find it very useful, particularly if students utilise ChatGPT to create the unique situation to which their test would react (Saunders 2023). Since the exam is custommade and student-generated, all of these requirements help to ensure its validity and guard against plagiarism and other academic integrity violations.

A few crucial tactics can be used by university personnel to create tests that discourage or limit students’ use of ChatGPT. One strategy is to design tests that demand students to show their ability to think critically, address challenges, and communicate. For instance, educators could create exams that require students to participate in group debates, presentations, or other interactive activities that require applying their knowledge and skills instead of just asking them to write an essay on a given subject. This would encourage critical thinking and self-directed learning while making it more challenging for students to finish tasks using ChatGPT or other LLMs.

Accurately citing and referencing other people’s work is expected of academic writing, including in-text citations and a list of references after the work. This promotes the legitimacy and dependability of the research and aids in acknowledging the original writers. Because ChatGPT and other LLMs might not have access to the same information sources or could need to be configured to style citations and references correctly, their output might need to contain proper referencing.

Developing open-ended tests and promoting creativity and originality are further tactics. Educators could, for instance, provide tests that require from students to formulate their research topics, to formulate and support their positions, or to come up with unique ideas and exercise critical thought. As a result, students may find it more challenging to use ChatGPT to finish their assignments. Lastly, to ensure that students are not using LLMs during assessments, the educator should employ proctored or real-time exams (Cotton et al. 2023:5 of 12).

There are numerous intriguing instances of ChatGPT’s possible applications, fantastic usage, and benefits in the academic setting. Among them are the following:

In light of these, educators could use ChatGPT in various ways to push students to acquire and hone their personal critical thinking abilities.

First, with questions like ‘How could you apply your past knowledge and abilities to address this new challenge?’ students are encouraged to think critically and solve difficulties by asking follow-up questions like ‘What strategies would you use to overcome any obstacles?’ Second, educators can routinely evaluate students’ strengths and areas for improvement by utilising ChatGPT to build quizzes and assessments that gauge students’ comprehension of the subject matter. Third, ChatGPT may create assignments that are differentiated based on the requirements and skills of each student, giving them progressively more complex assignments to push them. To provide students with high-quality examples, ChatGPT can also generate samples of excellent responses to tasks and assignments (Viljoen 2023).

Students in HE must actively engage with the material, critically reflect on their experiences and perspectives, relate the material to their personal and professional objectives, and synthesise new ideas and information to make meaning from it. Writing about their comprehension, applying what they have learnt to real-world situations, making visual aids or summaries, and discussing the learning material with others are some of the activities. Additionally, students’ perspectives can be widened and their knowledge deepened by actively searching and exposing them to various viewpoints.

LLMs significantly impact traditional teaching and learning approaches, and this development has led to a shift towards more experiential and student-centred learning approaches. Project-based learning is one example, which pushes students to use their knowledge and abilities to address challenges and complete projects in the real world. As a result, project-based learning gives students a context for their abilities and teaches them how to use them in practical situations.

Collaborative learning is another idea that is accomplished through teamwork and group projects. Students can use their knowledge and learn from one another in a dynamic and encouraging collaborative environment. Through practical and hands-on experiences like field excursions, job placements, and simulations, experiential learning (work-integrated learning) techniques enable students to apply their knowledge in real-world situations.

Finally, since it can be argued that technology can be used as a tool to facilitate learning and problem solving, incorporating technology into the educational process –such as gamification, AI, VR (virtual reality), and AR (augmented reality) – is an additional component in achieving this change (Viljoen 2023).

Responses from Questions Posted to Online Educators: Strategies for Online Assessment

The author sent five questions about ChatGPT and online assessment to 10 online educators. The answers can be summarised as follows:

Conclusion

Integrating advanced LLMs like ChatGPT into online assessments holds immense potential to revolutionise how we evaluate knowledge and skills. From adaptive assessments to automated grading and feedback, these models can enhance student and educator assessment experiences. However, careful attention must be given to addressing challenges such as bias, reliability, security, and privacy. Collaboration between researchers, educators, and policymakers is necessary to leverage the benefits of LLMs while upholding ethical standards and ensuring the fairness and validity of online assessments.

Instead of marking the conclusion of HE, ChatGPT has offered the sector, and society at large, a valuable opportunity. This is a chance to create innovative and inclusive approaches to teaching, learning, and assessment that correspond with these new understandings. Additionally, many scholars argue that educators should integrate new technologies into the educational experience. Universities and faculty members can and ought to do just that with Gen-AI technology. They should investigate it, identify its limitations, evaluate its potential applications within the relevant disciplines or teaching contexts, and engage in discussions about all these aspects with students, who are likely already familiar with the technology.

References

Ahmad, U. 2023. The potential of ChatGPT for personalized learning and

adaptive tutoring systems. Available at: https://www.linkedin.

com/pulse/potential-chatgpt-personalized-learning-adaptive

tutoring-usman-ahmad/ . (Accessed on 6 June 2023).

AlZu’bi, S., Mughaid, A., Quiam, F., & Samar H. 2023. Exploring the capabilities and limitations of ChatGPT and alternative big language models. Artificial Intelligence and Applications 20:1-16. https://doi.org/10.47852/bonviewAIA3202820

Bayat, M., Banihashem, SK., & Noroozi, O. 2022. The effects of collaborative reasoning strategies on improving primary school students’ argumentative decision-making skills. The Journal of Educational Research 115(6):349-358. https://doi.org/10.1080/002 20671.2022.2155602

Barber, M., Bird, L., Fleming, J., Titterington-Giles, E., Edwards, E., & Leyland, C. 2021. Gravity assist: Propelling higher education towards a brighter future. Available at: https://ofslivefs.blob. core.windows.net/files/Gravity%20assist/Gravity-assist-DTL finalforweb.pdf . (Accessed on 12 October 2023).

Barile, N. 2023. A guide to giving clear instructions to students (that they will actually follow). Available at: https://www.wgu.edu/ heyteach/article/guide-giving-clear-instructions-studentsthat-they-will-actually-follow2001.html#:~:text=After%20 giving%20them%20directions%2C%20ask%20your%20 students%20to,can%20help%20them%20comprehend%20 and%20complete%20the%20task. (Accessed on 7 June 2023).

Bekker, M. 2024. How academics can counter ‘AI thinks, therefore I am.’ University World News . Available at: https://www. universityworldnews.com/post.php?story=20240207185336305 . (Accessed on 8 February 2024).

Bhosale, SS., Pujari, V., Pujari, V., & Multani, Z. 2020. Advantages and disadvantages of artificial intelligence. Available at: https:// www.researchgate.net/publication/344584269_Advantages_ And_Disadvantages_Of_Artificial_Intellegence . (Accessed on 6 June 2023).

Biswas, S. 2023. Role of ChatGPT in computer programming. Mesopotamian Journal of Computer Science 2023:8-16. https://doi. org/10.58496/MJCSC/2023/002

Bommasani, R., Hudson, DA., Adeli, E., Altman, R., Arora, S., Von Arx, S., Bernstein, MS., Bohg, J., Bosselut, A., Brunskill, E., Brynjolfsson, E., Buch, S., Card, D., Castellon, R., Chatterji, N., Chen, A., Creel, K ., Davis, JQ., Demszky, D., Donahue, C., Doumbouya, M., Durmus, E., Ermon, S., Etchemendy, J., Ethayarajh, K., Fei- Fei, L., Finn, C., Gale, T., Gillespie, L., Goel, K., Goodman, N., Grossman, S., Guha, N., Hashimoto, T., Henderson, P., Hewitt, J., Ho, DE., Hong, J., Hsu, K., Huang, J., Icard, T., Jain, S., Jurafsky, D ., Kalluri, P., Karamcheti, S., Keeling, G., Khani, F., Khattab, O ., Koh, PW., Krass, M., Krishna, R., Kuditipudi, R., Kumar, A., Ladhak, F., Lee, M., Lee, T., Leskovec, J., Levent, I., Li, XL., Li, X ., Ma, T., Malik, A., Manning, CD., Mirchandani, S., Mitchell, E ., Munyikwa, Z., Nair, S., Narayan, A., Narayanan, D., Newman, B ., Nie, A., Niebles, JC., Nilforoshan, H., Nyarko, J., Ogut, G., Orr, L ., Papadimitriou, I., Park, JS., Piech, C., Portelance, E., Potts, C., Raghunathan, A., Reich, R., Ren, H., Rong, F., Roohani, Y., Ruiz, C ., Ryan, J., Ré, C., Sadigh, D., Sagawa, S., Santhanam, K., Shih, A ., Srinivasan, K., Tamkin, A., Taori, R., Thomas, AW., Tramèr, F., Wang, RE., Wang, W., Wu, B., Wu, J., Wu, Y., Xie, SM., Yasunaga, M., You, J., Zaharia, M., Zhang, M., Zhang, T., Zhang, X., Zhang, Y ., Zheng, L., Zhou, K., & Liang, P. 2021. On the opportunities and risks of foundation models. Available at: https://arxiv.org/ pdf/2108.07258.pdf . (Accessed on 7 June 2023).

Brown, TB., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., Agarwal, S., Herbert-Voss, A., Krueger, G., Henighan, T., Child, R., Ramesh, A., Ziegler, DM., Wu, J., Winter, C., Hesse, C., Chen, M., Sigler, E., Litwin, M., Gray, S., Chess, B., Clark, J., Berner, C., McCandlish, S., Radford, A., Sutskever, I., & Amodei, D. 2020. Language models are few-shot learners. Available at: https://arxiv.org/ pdf/2005.14165.pdf . (Accessed on 6 June 2023).

Cascella, M., Montomoli, J., Bellini, V., & Bignami, E. 2023. Evaluating the feasibility of ChatGPT in healthcare: An analysis of multiple clinical and research scenarios. Journal of Medical Systems 47(33):1-5. https://doi.org/10.1007/s10916-023-01925-4

Celik, I., Dindar, M., Muukkonen, H., & Järvelä, S. 2022. The promises and challenges of artificial intelligence for teachers: A systematic review of research. TechTrends 66:616-630. https://doi. org/10.1007/s11528-022-00715-y

Cotton, DRE., Cotton, PA., & Shipway, JR. 2023. Chatting and cheating: Ensuring academic integrity in the era of ChatGPT. Innovations in Education and Teaching International . 12 pages. https://doi. org/10.35542/osf.io/mrz8h

Chowdhury, NA. & Rahman, S. 2023. A brief review of ChatGPT: Limitations, challenges and ethical-social implications. Available at: https://www.researchgate.net/publication/368397881_A_ brief_review_of_ChatGPT_Limitations_Challenges_and_ Ethical-Social_Implications . (Accessed on 6 June 2023).

Dehouche, N. 2021. Plagiarism in the age of massive generative pretrained transformers (GPT-3). Ethics in Science and Environmental Politics 21:17-23. https://doi.org/10.3354/esep00195

Dheda, G. 2023. What does GPT stand for in Chat GPT: Everything you need to know. Available at: https://openaimaster.com/what does-gpt-stand-for-in-chat-gpt/ . (Accessed on 6 June 2023).

Farrokhnia, M., Banihashem, SK., Noroozi, O., & Wals, A. 2023. A SWOT analysis of ChatGPT: Implications for educational practice and research. Innovations in Education and Teaching International. 15 pages. https://doi.org/10.1080/14703297.2023.2195846

Gao, J. 2021. Exploring the feedback quality of an automated writing evaluation system Pigai. International Journal of Emerging Technologies in Learning 16(11):322-330. https://doi.org/10.3991/ ijet.v16i11.19657

Gewirtz, D. 2023. How to make ChatGPT provide sources and citations. Available at: https://www.zdnet.com/article/how-to-make chatgpt-provide-sources-and-citations/ . (Accessed on 4 October 2023).

Gpt Generative Pretrained Transformer, Thunström, AO., & Steingrimsson, S. 2022. Can GPT-3 write an academic paper on itself, with minimal human input? 7 pages. Available at: https:// hal.archives-ouvertes.fr/hal-03701250/document . (Accessed on 11 October 2023).

Hapsari, IP. & Wu, TT. 2022. AI chatbots learning model in English speaking skill: Alleviating speaking anxiety, boosting enjoyment, and fostering critical thinking. In Huang, YM., Cheng, SC., Barroso, J., & Sandnes, FE. (Eds): Innovative technologies and learning, 444-453. ICITL 2022. Lecture Notes in Computer Science, vol 13449. Cham: Springer. https://doi.org/10.1007/978 3-031-15273-3_49

Hoyos, A. 2023. Unpacking ChatGPT: The pros and cons of AI’s hottest language model. Available at: https://www.ie.edu/insights/ articles/unpacking-chatgpt-the-pros-and-cons-of-ais hottest-language-model/ . (Accessed on 12 October 2023).

Jia, Q., Cui, J., Xiao, Y., Liu, C., Rashid, P., & Gehringer, E. 2021. ALL-INONE: Multi-task learning BERT models for evaluating peer assessments. Available at: https://arxiv.org/pdf/2110.03895.pdf. (Accessed on 12 October 2023).

Jia, F., Sun, D., Ma, Q., & Looi, CK. 2022. Developing an AI-based learning system for L2 learners’ authentic and ubiquitous learning in English language. Sustainability 14(23). 15527. 18 pages. https:// doi.org/10.3390/su142315527

Karthikeyan, C. 2023. Literature review on pros and cons of ChatGPT implications in education. International Journal of Science and Research 12(13):283-291. https://doi.org/10.21275/SR23219122412

Leadership Workshop. 2012. Faculty evaluation rubrics. University of Washington Center for Instructional Change. Available at: https://advance.washington.edu/sites/advance/files/1a.%20 Faculty%20Evaluation%20Rubrics.pdf#:~:text=Rubrics%20 are%20often%20used%20in%20teaching%20to%20 assess,useful%20for%20evaluating%20faculty%20 candidates%20and%20faculty%20performance. (Accessed on 7 June 2023).

Lewis, A. 2022. Multimodal large language models for inclusive collaboration learning tasks. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Student Research Workshop, Washington , 202-210. Available at: https:// aclanthology.org/2022.naacl-srw.26.pdf . (Accessed on 12 October 2023).

Li, C. & Xing, W. 2021. Natural language generation using deep learning to support MOOC learners. International Journal of Artificial Intelligence in Education 31(2):186-214. https://doi.org/10.1007/ s40593-020-00235-x

Lin, F., Cao, W., Zhang, W., Chen, L., Hong, Y., Xu, T., & Tan, C. 2022. Knowledge-enhanced retrieval: A scheme for question answering. Conference paper. Available at: https://link.springer. com/chapter/10.1007/978-981-19-0713-5_12 . (Accessed on 12 October 2023).

MacGregor, K. 2023. ChatGPT – calm your inner Luddite, keep your inner sceptic. University World News . Available at: https://www. universityworldnews.com/post.php?story=20230302062634124 . (Accessed on 2 March 2023).

Marr, B. 2023. The top 10 limitations of ChatGPT. Available at: https:// www.forbes.com/sites/bernardmarr/2023/03/03/the-top-10 limitations-of-chatgpt/ . (Accessed on 6 June 2023).

Messer, M. 2022. Grading programming assignments with an automated grading and feedback assistant. Conference paper. Available at: https://link.springer.com/ chapter/10.1007/978-3-031-11647-6_6 . (Accessed on 7 June 2023).

Mizumoto, A. & Eguchi, M. 2023. Exploring the potential of using an AI language model for automated essay scoring. SSRN Electronic Journal . 40 pages. https://doi.org/10.2139/ssrn.4373111

Perez, S., Massey-Allard, J., Butler, D., Ives, J., Bonn, D., Yee, N., & Roll,

I. 2017. Identifying productive inquiry in virtual labs using sequence mining. In André, E., Baker, R., Hu, X., Rodrigo, MMT., & Du Boulay, B. (Eds.): Artificial intelligence in education, 287-298. Proceedings of the 18th International Conference, AIED 2017. Wuhan, China, 28 June to 1 July 2017. Available at: https://files.eric.ed.gov/ fulltext/ED617869.pdf . (Accessed on 7 June 2023).

Qadir, J. 2022. Engineering education in the era of ChatGPT: Promise and pitfalls of generative AI for education. Available at: https:// www.researchgate.net/publication/366712815_Engineering_ Education_in_the_Era_of_ChatGPT_Promise_and_Pitfalls_ of_Generative_AI_for_Education . (Accessed on 7 June 2023).

Rudolph, J., Tan, S., & Tan, S. 2023. ChatGPT: Bullshit spewer or the end of traditional assessments in higher education? Journal of Applied Learning & Teaching 6(1):342-362. https://doi.org/10.37074/ jalt.2023.6.1.9

Saunders, S. 2023. Rather than ban generative AI, universities must learn from the past. University World News . Available at: https://www. universityworldnews.com/post.php?story=20230302062634124 . (Accessed on 6 March 2023).

Sharma, Y. 2024. Generative AI prompts new areas of university research. Available from: https://www.universityworldnews. com/post.php?story_20240129205809349&utm_ source=newsletter&utm_medium=email&utm_ campaign=SPNL5008 . (Accessed on 3 February 2024).

Sümer, O. 2023. Artificial intelligence chatbots, how do they work? For which purposes can they be used? What are the benefits and potential pitfalls? Available at: https://www.omersumer.com/ personal-writings/2023-03-21-ChatGPT-and-LLMs/index. html . (Accessed on 7 June 2023).

Viljoen, J. 2023. AI in higher education – a tool for better learning? University World News . Available at: https://www. universityworldnews.com/post.php?story=20230302062634124 . (Accessed on 2 March 2023).

Wong, C. 2023. Automatic grading with Azure OpenAI Services ChatGPT virtual assistant. Available at: https://techcommunity.microsoft. com/t5/educator-developer-blog/automatic-grading-with azure-openai-services-chatgpt-virtual/ba-p/3811231 . (Accessed on 6 June 2023).

Zawacki-Richter, O., Marín, VI., Bond, M., & Gouverneur, F. 2019. Systematic review of research on artificial intelligence applications in higher education – where are the educators? International Journal of Educational Technology in Higher Education 16(1):1-27. https://doi.org/10.1186/s41239-019-0171-0

Zhai, X. 2022. ChatGPT user experience: Implications for education. SSRN Electronic Journal . 18 pages. Available at: https://ssrn.com/ abstract=4312418 . http://dx.doi.org/10.2139/ssrn.4312418

Introduction

AI is not a recent concept, as Isaac Asimov has already made his infamous robotic invention based on the Three laws of robotics and the science fiction book, I Robot (1950). Alan Turing extended AI research in the 1950s, which led UNESCO (United Nations Educational, Scientific, and Cultural Organisation) to

recommend ethical considerations for using AI tools to prevent abuse, fraud, and security risks. Based on the principles of AI

ethics, different AI tools were developed to minimise academic integrity risks. Those early years of exploring AI in HE (higher

education) opened many opportunities and challenges to

the sector. The rapid development of AI research is having a

profound impact on HE. AI solutions have great potential to

speed up progress towards the SDGs (sustainable development

goals) and the Africa Agenda 2063 vision.

Within the past decade, enormous strides have been made related to the effect of AI on education. According to Holmes, Bialik, and Fadel (2023:633), the application of Gen-AI (generative artificial intelligence) tools enhances education

and has grown exponentially. Scholarly works have reported how Gen-AI affects tuition, research, assessment, ethics, and

professional development in the HE sector (Zawacki-Richter,

Marín, Bond, & Gouverneur 2019:23). On the other hand, Nemorin, Vlachidis, Ayerakwa, and Andriotis (2023:39) doubt

the claims about the revolutionary potential of AI in education, stating that they are primarily based on conjecture, speculation, and optimism. Some scholars discuss whether Gen-AI tools can track learning progress in different situations and skills. However, at the time of exploring academic staff’s awareness of Gen-AI conversational tools in teaching and learning (Holmes

et al. 2023:633), there have been studies conducted at IHEs (institutions of higher education) globally (Smolansky, Cram, Raduescu, Zeivots, Huber, & Kizilcec 2023; Peres, Schreier,

Schweidel, & Sorescu 2023) but to a lesser extent in the African context (Baidoo-Anu & Ansah 2023; Mhlanga 2023; Van Wyk,

Adarkwah, & Amponsah 2023:216). In November 2022, a

tech company launched ChatGPT 3.5. This sophisticated AI-

based conversational tool created a global movement with the

characteristics of the infamous updated version, ChatGPT 3.5, which the tech company OpenAI developed. This app created a buzz around a Gen-AI tool that can mimic human responses to text prompts. ChatGPT grew faster than any other social media and became the fastest-growing app, but similar LLMs (large

language model) emerged, like Claude2 and Llama2, as chatbots

(Rudolph, Tan, & Tan 2023).

ChatGPT’s advances have emphasised the importance of researching it in education. At a South African university, the author studied academic experiences with ChatGPT, a Gen-AIbased conversation tool. Some participants were exposed to and used Gen-AI tools like ChatGPT 3.5 to experiment with teaching and learning activities. This inquiry generated the following research questions:

This study made three significant contributions to Gen-AIbased LLM research. Using Gen-AI in teaching is being studied, and it is contributing to that research. The study expands our understanding of Gen-AI, particularly ChatGPT 3.5 as a chatbotbased LLM in the faculty of education. Third, one should design strategies and create an awareness of detector tools to prevent academic dishonesty. Practical implications included policy revisions for ethics, tuition, and research. There is also a need for continuous professional development in the faculty of education.

Literature Review

Methodology

Since universities became known with ChatGPT 3.5, they started with critical conversations about Gen-AI in education. For example, throughout 2023, at this mentioned African university, academics were introduced to online webinars, face-to-face seminars, college discussion forums, and virtual conferences, centred around Gen-AI tools, in particular ChatGPT 3.5 in HE. In one of the virtual Microsoft Teams videoconferencing sessions on ChatGPT 3.5 and detectors, an invitation was sent to academics to participate in an awareness study related to Gen-AI conversational tools. Based on this exposure, the participants agreed and signed a consent application to accept being a part of the study. Based on the notion of phenomenology, African scholars’ phenomenology is the science of experiences-as-lived and both a theory and a method used to explore academic awareness and experiences with Gen-AI tools by academics (Koopman & Koopman 2018:16; Van Wyk 2022:124). Various research methods have been used to measure Gen-AI tools, but the chosen method was IPA (interpretative phenomenology analysis) for this study. Smith and Osborn (2008:67) argue that an IPA approach is conducted to create a narrative account for each participant through verbatim statements, views, and beliefs. For this study, the IPA is appropriate for exploring four academics’ lived experiences and how specific strategies to prevent cheating and academic dishonesty were applied in teaching and learning. The purposive sampling comprised academics. Pseudonym names were used, such as lecturers, senior lecturers, associate professors, and full professors, to protect the identity of participants. Dates and times were scheduled on Microsoft Teams to conduct the interviews. The interviews conducted on Microsoft Teams were transcribed. We then downloaded and secured the recordings as part of data management. An online link was sent to participants to verify the accuracy of recordings, transcripts, and questions. After a week, if participants agreed, an e-mail (no names to protect identification, codes as 12345@unisa.ac.za ) was sent to the study’s author. The participants stated in their e-mailed responses that they agreed that the recordings, data generated, and transcripts were a true reflection. The e-mail served as proof of their participant validation and consent.

The data exploitation approach for conducting the IPA approach as provided by Hycner (1999) indicates how the simplified ‘explicitation’ strategy as an ‘investigation of constituents of the phenomenon while keeping the context of the whole’ (Hycner 1999:161) is used. The extracts were analysed and themes were identified as guided by the IPA analysis process (Hycner 1999:148; Creswell 2013:slide 23; Nowell, Norris, White, & Moules 2017). The ethical clearance for the study was granted before it started (reference 2020/08/12/90178912/19/AM).

Findings

The results confirm that researchers conducted minimal research on ChatGPT as an AI conversational tool. Academics were aware of the potential benefits and drawbacks of ChatGPT for teaching and learning. Strategies were developed to prevent academic cheating in AI-based learning.

Discussion of Findings

In this study, the participants were of the view that Gen-AI tools could be an advantage in teaching and learning. Chatbots support students with case studies, problem-solving, project-based learning, and academic writing using AI-conversational tools (Kaplan-Rakowski et al. 2023:30). Participants echoed sentiments of the usefulness of Gen-AI tools in promoting advancing teaching and learning experiences (Megahed, Chen, Ferris, Knoth, & Jones-Farmer 2023:27 of 30; Castillo et al. 2023:11). According to Participant 8, ‘ChatGPT was very useful in developing academic writing skills for postgraduate students.’ Another issue that emerged from the interviews is that the disruptive Gen-AI conversational tools will likely change the face of HE immensely. Mohamed (2024:3208) reports that the lecturers viewed ChatGPT as efficient in teaching English Foreign Language students. Students were concerned about the cost and availability of AI tools for those who are disadvantaged, as companies were investing in them for profit.

Participants claimed that it is vital to be aware of and consider ethics when using AI chatbots. They felt academics must be cognisant of the speed of the emergence of new Gen-AI tools since the launch of ChatGPT 3.0 in 2022. Studies concur that the ethics of using Gen-AI tools by students is a major issue that needs to be addressed (Baek & Kim 2023; Eke 2023:4 of 4; Cotton et al. 2023:12). Certain studies are expressing the fear of ethical implications, as cheating, plagiarism, and copyright infringement have ethical implications for education (Cooper 2023:449). The participants concurred that ethics is at the heart of preventing and detecting cheating or academic dishonesty. There were already debates about ethical issues like academic dishonesty and cheating due to using Gen-AI tools. There are AI-detection software to detect academic dishonesty, but these are expensive for HE (Oravec 2023:213).

IHEs must update policies and educate staff and students to prevent cheating with Gen-AI tools (Dwivedi et al. 2023:61 of 63). According to the participants, these are critical strategies against academic dishonesty.

Certain measures are proposed to prevent academic cheating using AI in teaching and learning. Participants believed it necessary to prevent academic dishonesty and adopted measures to combat fraud and cheating. Students are allowed to present case studies, research, and problem-based projects as a group and report the results (Firaina & Sulisworo 2023:39). Participant 9 proposed an ‘alternative assessment by creating project-based learning tasks for students to plan, implement, and present their final projects.’ Studies advocate the use of AI chatbot strategies to prevent academic dishonesty (Ali 2023:4; Cotton et al. 2023:12). Several participants believed it is vital to expose students to detector software as a way to discourage cheating. Participants opined that to prevent cheating among students, awareness, adoption, and design of authentic learning experiences will ultimately change the usefulness of chatbots. Webinars on Gen-AI conversational tools are a valuable strategy to increase awareness. According to Castillo et al. (2023:10), discussion forums can be used as an effective, active method to create critical conversations about Gen-AI detector software to inculcate a sense of student adoption of detector tools.

Gen-AI tools had exponentially grown globally, so education had to change their teaching and learning. According to Castillo et al. (2023:16), ChatGPT had a significant impact on student learning. Scholars posit that soft skills like social interaction, creativity, curiosity, and reflective thinking are vital for futuristic workforce placement. Studies reported that students are empowered with soft skills to succeed in an AI-dominated workplace (Baidoo-Anu & Ansah 2023:57; Mukherjee 2023:124; Reilly 2018:52; Toli 2023:79). Participant 5 argued that ‘soft skills are important to prepare students for future employability. Universities must rethink and reimage the relevant academic programmes to equip students with problem-solving, critical thinking, emotional intelligence, communication, interpersonal relations, and technical skills to be relevant for employability in an AI-dominated landscape.’

Conclusion

Since the launch of the first LLM, several scholarly publications have reported benefits, but some scholars have raised concerns about ethics and morality. These LLMs have brought to light a significant pedagogical change in HE. This exploratory study revealed that academic staff created an awareness of the ethics of ChatGPT. Some participants raised a concern about whether ChatGPT is capable of generating accurate information because the world is full of inaccurate knowledge due to many chatbots. Participants believed that specific strategies could prevent academic dishonesty and cheating. The author of this study has suggested some strategies to prevent cheating with Gen-AI tools. Academics must redesign authentic context-based assessment tasks, such as specific case studies, problem-solving activities, and project-based learning tasks, and academic writing skills.

As suggested by participants, these strategies create hands-on learning tasks, introduce students to Gen-AI detection tools, and create authentic learning opportunities. The participants suggested that education faculties need to update AI policies, provide guidelines and educate students about AI literacy skills and AI detector tools for the ethical use of Gen-AI tools in teaching and learning. Furthermore, sentiments revealed that education must rethink and reimagine existing curricula to strategise by empowering student educators with soft skills to succeed in an AI-dominated workplace.

This study made significant contributions to advance an understanding of Gen-AI-based LLM research. This research indicates how Gen-AI can stop cheating in academia. It is suggested that awareness needs to be created using discussion forums and online panels as critical conversations about detector software like GPTZero. This study extends the theoretical knowledge of Gen-AI, in particular, an awareness of the ‘truth or fake’ of Gen-AI. The author recommended modifications to the education of educators, covering ethics, tuition, research, assessment, and professional development. Futuristic inquiry is to be conducted to rethink curricula to enhance the soft skills of student educators in preparing for an AI-dominated workplace.

References

Abawi, LA. 2015. Inclusion ‘from the gate in:’ Wrapping students with personalised learning support. International Journal of Pedagogies and Learning 10(1):47-61. https://doi.org/10.1080/22040552.201 5.1084676

Ali, F. 2023. Let the devil speak for itself: Should ChatGPT be allowed or banned in hospitality and tourism schools? Journal of Global Hospitality and Tourism 2(1):1-6. https://doi.org/10.5038/2771 5957.2.1.1016

Antonenko, P. & Abramowitz, B. 2023. In-service teachers’ (mis) conceptions of artificial intelligence in K-12 science education. Journal of Research on Technology in Education 55(1):64-78. https://doi.org/10.1080/15391523.2022.2119450

Baidoo-Anu, D. & Ansah, LO. 2023. Education in the era of generative artificial intelligence (AI): Understanding the potential benefits of ChatGPT in promoting teaching and learning. Journal of AI 7(1):52-62. http://dx.doi.org/10.2139/ssrn.4337484

Baek, TH. & Kim, M. 2023. Is ChatGPT scary good? How user motivations affect creepiness and trust in generative artificial intelligence. Telematics and Informatics 83. 102030. https://doi.org/10.1016/j. tele.2023.102030

Canzonetta, J. 2021. Repurposing plagiarism detection services for responsible pedagogical application and (in)formative assessment of source attribution practices. Assessing Writing 50. 100563. https://doi.org/10.1016/j.asw.2021.100563

Castillo, AGR., Silva, GJS., Arocutipa, JPF., Berrios, HQ., Rodriguez, MAM., Reyes, GY., Lopez, HRP., Herbert, RMVT., Rivera, VH., & Arias-Gonzáles, JL. 2023. Effect of Chat GPT on the digitized learning process of university students. Journal of Namibian Studies: History Politics Culture 33:1-15. https://doi.org/10.59670/jns.v33i.411

Chaudhry, MA. & Kazim, E. 2021. Artificial intelligence in education (AIEd): A high-level academic and industry note. AI and Ethics 2:1-16. https://doi.org/10.1007/s43681-021-00074-z

Chan, CKY. & Hu, W. 2023. Students’ voices on generative AI: Perceptions, benefits, and challenges in higher education. International Journal of Educational Technology in Higher Education 20. 43. 18 pages. https://doi.org/10.1186/s41239-023-00411-8

Cooper, G. 2023. Examining science education in chatbot: An exploratory study of generative artificial intelligence. Journal of Science Education and Technology 32(3):444-452. https://doi.org/10.1007/ s10956-023-10039-y

Cotton, DRE., Cotton, PA., & Shipway, JR. 2023. Chatting and cheating: Ensuring academic integrity in the era of ChatGPT. Innovations in Education and Teaching International 61(2):1-12. https://doi.org/10 .1080/14703297.2023.2190148

Cope, B., Kalantzis, M., & Searsmith, D. 2021. Artificial intelligence for education: Knowledge and its assessment in AI-enabled learning ecologies. Educational Philosophy and Theory 53(12):1229-1245. https://doi.org/10.1080/00131857.2020.1728732

Creswell, JW. 2013. Steps in conducting a scholarly mixed methods study. Discipline-Based Education Research Group. Paper 48. PowerPoint: 52 slides.

Dwivedi, YK., Kshetri, N., Hughes, L., Slade, EL., Jeyaraj, A., Kar, AK., Baabdullah, AM., Koohang, A., Raghavan, V., Ahuja, M., & Albanna, H. 2023. ‘So what if ChatGPT wrote it?’ Multidisciplinary perspectives on opportunities, challenges and implications of generative conversational AI for research, practice and policy. International Journal of Information Management 71. 102642. 63 pages. https://doi.org/10.1016/j. ijinfomgt.2023.102642

Eke, DO. 2023. ChatGPT and the rise of generative AI: A threat to academic integrity? Journal of Responsible Technology 13. 100060. 4 pages. https://doi.org/10.1016/j.jrt.2023.100060

Firaina, R. & Sulisworo, D. 2023. Exploring the usage of ChatGPT in higher education: Frequency and impact on productivity. Buletin Edukasi Indonesia 2 (01):39-46. https://doi.org/10.56741/bei.v2i01.310

Griffith, E. & Metz, C. 2023. ‘Let 1,000 flowers bloom:’ AI funding frenzy escalates. The New York Times . Available at: https:// www.nytimes.com/2023/03/14/technology/ai-funding-boom. html?action=click&module=RelatedLinks&pgtype=Article . (Accessed on 11 January 2024).

Holmes, W., Bialik, M., & Fadel, C. 2023. Artificial intelligence in education. In Stückelberger, C. & Duggal, P. (Eds.): Data ethics: building trust: how digital technologies can serve humanity, 621-653. Geneva: Globethics. https://doi.org/10.58863/20.500.12424/4276068

Huang, W., Hew, KF., & Fryer, LK. 2022. Chatbots for language learning – are they really useful? A systematic review of chatbot-supported language learning. Journal of Computer Assisted Learning 38(1):237-257. https://doi.org/10.1111/jcal.12610

Huallpa, JJ. 2023. Exploring the ethical considerations of using Chat GPT in university education. Periodicals of Engineering and Natural Sciences 11(4):105-115. https://doi.org/10.21533/pen.v11.i4.200

Hycner, RH. 1999. Some guidelines for the phenomenological analysis of interview data. In Bryman, A. & Burgess, RG. (Eds.): Qualitative research 3, 143-164. London: Sage.

Kaplan-Rakowski, R., Grotewold, K., Hartwick, P., & Papin, K. 2023. Generative AI and teachers’ perspectives on its implementation in education. Journal of Interactive Learning Research 34(2/3):1333.

Kelly, A., Sullivan, M., & Strampel, K. 2023. Generative artificial intelligence: University student awareness, experience, and confidence in use across disciplines. Journal of University Teaching & Learning Practice 20(6). 12. 16 pages. https://doi. org/10.53761/1.20.6.12

Koopman, O. & Koopman, KJ. 2018. The body as blind spot: Towards lived experience and a body-specific philosophy in education. Education as Change 22(3):1-16. https://doi.org/10.25159/1947 9417/1880

Leaver, T. & Srdarov, S. 2023. ChatGPT isn’t magic: The hype and hypocrisy of generative artificial intelligence (AI) rhetoric. M/C (Media & Culture) Journal 26(5). 11 pages. https://doi.org/10.5204/ mcj.3004

Ledwith, A. & Risquez, A. 2008. Using anti-plagiarism software to promote academic honesty in the context of peer reviewed assignments. Studies in Higher Education 33(4):371-384. https:// doi.org/10.1080/03075070802211562

Li, KC. & Wong, BTM. 2023. Artificial intelligence in personalised learning: A bibliometric analysis. Interactive Technology and Smart Education 20(3):422-445. https://doi.org/10.1108/ITSE-01-2023 0007

Lim, WM., Gunasekara, A., Pallant, JL., Pallant, JI., & Pechenkina, E. 2023. Generative AI and the future of education: Ragnarök or reformation? A paradoxical perspective from management educators. The International Journal of Management Education 21(2). 100790. 13 pages. https://doi.org/10.1016/j. ijme.2023.100790

Liu, Z., Yao, Z., Li, F., & Luo, B. 2023. Check me if you can: Detecting ChatGPT-generated academic writing using CheckGPT. arXiv:2306.05524. 23 pages. https://doi.org/10.48550/ arXiv.2306.05524

Lodge, JM., Thompson, K., & Corrin, L. 2023. Mapping out a research agenda for generative artificial intelligence in tertiary education. Australasian Journal of Educational Technology 39(1):1-8.

Luan, H., Geczy, P., Lai, H., Gobert, J., Yang, SJH., Ogata, H., Baltes, J., Guerra, R., Li, P., & Tsai, C-C. 2020. Challenges and future directions of big data and artificial intelligence in education. Frontiers in Psychology 11. 580820. 11 pages. https://doi. org/10.3389/fpsyg.2020.580820

Mhlanga, D. 2023. Open AI in education, the responsible and ethical use of ChatGPT towards lifelong learning. In Mhlanga, D. (Ed.): FinTech and Artificial Intelligence for Sustainable Development: The role of smart technologies in achieving development goals, 387-409. Cham: Palgrave Macmillan. https://doi.org/10.1007/978-3-031 37776-1_17

Megahed, FM., Chen, YJ., Ferris, JA., Knoth, S., & Jones-Farmer, LA. 2023. How generative AI models such as ChatGPT can be (mis)used in SPC practice, education, and research? An exploratory study. Quality Engineering . arXiv:2302.109161v1. 30 pages. https://arxiv. org/pdf/2302.10916.pdf .

Mohamed, AM. 2024. Exploring the potential of an AI-based chatbot (ChatGPT) in enhancing English as a foreign language (EFL) teaching: Perceptions of EFL faculty members. Education and Information Technologies 29: 3195-3217. https://doi.org/10.1007/ s10639-023-11917-z

Mukherjee, DV. 2023. At the edge of tomorrow: Unleashing human potential in the AI era. Chennai: Notion Press.

Nemorin, S., Vlachidis, A., Ayerakwa, HM., & Andriotis, P. 2023. AI hyped? A horizon scan of discourse on artificial intelligence in education (AIED) and development. Learning, Media and Technology 48(1):38-51. https://doi.org/10.1080/17439884.2022.2095568

Nowell, LS., Norris, JM., White, DE., & Moules, NJ. 2017. Thematic analysis: Striving to meet the trustworthiness criteria. International Journal of Qualitative Methods 16(1):1-13. https://doi. org/10.1177/1609406917733847

Oravec, JA. 2023. Artificial intelligence implications for academic

cheating: Expanding the dimensions of responsible human-AI collaboration with ChatGPT. Journal of Interactive Learning Research 34(2):213-237.

Peres, R., Schreier, M., Schweidel, D., & Sorescu, A. 2023. On ChatGPT and beyond: How generative artificial intelligence may affect research, teaching, and practice. International Journal of Research in Marketing 40(2):269-275. https://doi.org/10.1016/j. ijresmar.2023.03.001

Perkins, M. 2023. Academic integrity considerations of AI large language models in the post-pandemic era: ChatGPT and beyond. Journal of University Teaching & Learning Practice 20(2):1-23. https://doi. org/10.53761/1.20.02.07

Popenici, SA. & Kerr, S. 2017. Exploring the impact of artificial intelligence on teaching and learning in higher education. Research and Practice in Technology Enhanced Learning 12(1):1-13. https://doi. org/10.1186/s41039-017-0062-8

Reilly, P. 2018. The impact of artificial intelligence on the HR function. IES Perspectives Member Paper 142, 41-58. Available at: https:// www.employmentstudies.co.uk/system/files/resources/files/ . (Accessed on 25 April 2024).

Rudolph, J., Tan, S., & Tan, S. 2023. War of the chatbots: Bard, Bing Chat, ChatGPT, Ernie and beyond. The new AI gold rush and its impact on higher education. Journal of Applied Learning and Teaching 6(1):364-389. https://doi.org/10.37074/jalt.2023.6.1.23

Sallam, M. 2023. ChatGPT utility in healthcare education, research, and practice: Systematic review on the promising perspectives and valid concerns. Healthcare 11(6). 887. 20 pages. https://doi. org/10.3390/healthcare11060887

Shang, H-F. 2019. An investigation of plagiarism software use and awareness training on English as a foreign language (EFL) students. Journal of Computing in Higher Education 31(1):105-120. https://doi.org/10.1007/s12528-018-9193-1

Smith, J A. & Osborn, M. 2008. Interpretative phenomenological analysis. In Smith, J. (Ed.): Qualitative psychology: A practical guide to research methods, 53-80. London: Sage.

Smolansky, A., Cram, A., Raduescu, C., Zeivots, S., Huber, E., & Kizilcec, RF. 2023. Educator and student perspectives on the impact of generative AI on assessments in higher education. Proceedings of the Tenth ACM Conference on Learning @ Scale, 378-382. Available at: https://www.researchgate.net/publication/372482047_ Educator_and_Student_Perspectives_on_the_Impact_of_ Generative_AI_on_Assessments_in_Higher_Education . (Accessed on 25 October 2023).

Taecharungroj, V. 2023. ‘What can ChatGPT do?’ Analyzing early reactions to the innovative AI chatbot on Twitter. Big Data and Cognitive Computing 7(1). 35. 10 pages. https://doi.org/10.3390/ bdcc7010035

Toli, L. 2023. The impact of AI technology on higher education in India: Revolutionizing learning and empowering students. International Journal of Multidisciplinary Educational Research 12(10):76-80. http://ijmer.in.doi./2023/12.10.92

Van Wyk, MM., Adarkwah, MA., & Amponsah, S. 2023. Why all the hype about ChatGPT? Academics’ views of a chat-based conversational learning strategy at an open distance e-learning institution. Open Praxis 15(3):214-225. https://doi.org/10.55982/ openpraxis.15.3.563

Van Wyk, MM. 2022. Economics student teachers’ experiences-as-lived in a flipped pedagogical approach during crisis times. African Perspectives of Research in Teaching and Learning 6(3):121-139.

Yang, W. 2022. Artificial intelligence education for young children: Why, what, and how in curriculum design and implementation. Computers and Education: Artificial Intelligence 3. 100061. 7 pages. https://doi.org/10.1016/j.caeai.2022.100061

Zawacki-Richter, O., Marín, VI., Bond, M., & Gouverneur, F. 2019. A systematic review of research on artificial intelligence applications in higher education – where are the educators? International Journal of Educational Technology in Higher Education 16(1):1-27. https://doi.org/10.1186/s41239-019-0171-0