• Daystar University
• 10 Nov, 2023

By Rev. Mutinda Musyimi, Dean of Students

The 2020COVID-19 pandemic launched the unavoidable global epidemic of loneliness into existence. Smart devices, unintentionally, have created a connectivity paradox: the previously disconnected have become more connected and the previously connected have become more disconnected. Consequently, every institution of higher learning has to deal with a generation that has shifted their values of connection, a generation that is aware that artificial relationships are neither good nor desirable for them, yet, a generation without social stamina on how to deal with this social predicament.

New Plague

Barely three years after the COVID-19 pandemic, institutions of higher learning are faced with a new sort of plague, one that threatens

our minds more than our bodies, one that is creating an infodemic. The Large Language Models (LLMs) have recently gathered attention with the release of a user-centered chatbot. You have probably heard of or already used ChatGPT, yes; even one medical insurance underwriter has ChatGPT on their billboards! This artificial intelligence (AI) chatbot was developed by Open AI and launched in November 2022. By January 2023, ChatGPT reached over 100 million users, and currently boasts over 1.5 billion visitors per month, making it the fastest growing consumer application to date.

The impact of ChatGPT is huge for the general public and the academia as well, with many students using the chatbot to write their research work, and some papers even listing ChatGPT as an author. ChatGPT can be used by students and researchers to conduct research and create important articles, including suggesting relevant titles for research articles, writing drafts and expressing convoluted social science and scientific concepts. But as we celebrate the birth of this more fickle than bouncy baby, there are alarming ethical and practical

challenges emerging from the use of LLMs. To what extent, for example, can generative AI conceptualize an emergent, not well-described field, since it relies on previous human inputs? AI technologies like ChatGPT can inherit biases from the data they are trained on, which may result in biased or illusory content generation that could influence teaching and learning undesirably.

Turnitin

Would it be correct to say we are largely defenseless against this novel threat to human intelligence and academic integrity? For example, the inability to accurately detect artificial- intelligence-produced text still remains an unresolved issue. True, several GPT detectors are already available, with Turnitin, the main company employed by colleges to detect plagiarism in student submissions, breathlessly trying to catch up. A potential concern related to the emergence of LLMs is the submissiveness in following users’ commands. But the most significant handicap of ChatGPT is that the information it compiles is not always accurate. After all, academic progress is contingent on sharing accurate and verified information. This drawback is especially detrimental in scientific research and publishing. Presenting incorrect data in a scientific setting

Congratulations to our Dons for acquiring new hats! Daystar University Congratulates the Dean, School of Communication, Prof. Levi Obonyo, upon his appointment to Full Professor of Media and Communication, of Daystar University. In addition, Prof. Obonyo has been appointed as the Associate Editor: Annals of International Communication Association (ICA). Dr. Rose Mwangi-Mburu PhD, Communication & Media Studies, Kenyatta University Dr. Joab Namai Obuchere PhD, Education Administration & Planning,Catholic University of Eastern Africa Dr. Myra Ohaga-Gitonga PhD, Information Technology, Kibabii University Dr. Sylvia Wakene PhD, International Relations, United States International University Africa Dr. Julie Muraya PhD, Communication Studies Moi University carries a great risk of harm, and as it stands, ChatGPT cannot be relied upon to provide correct facts or produce reliable inferences and references.

Problem Based Learning

Daystar University has been keen on using Problem-Based Learning [PBL]. This is a student-centered teaching method in which complex real-world problems are used to promote student learning of concepts and principles as opposed to direct presentation of facts and concepts. The use of AI language models may raise ethical concerns around academic integrity and plagiarism. But beyond these, there ought to be a growing concern among educators and experts that the convenience and speed of ChatGPT in the classroom could lead to a lack of critical thinking, problem-solving skills, and creativity in the educational process. Subjects such as literature, history, and philosophy require students to engage with complex ideas and perspectives, and to conclusively

develop their own arguments and elucidations based on investigation and evidence. Shouldn’t we be concerned that ChatGPT might lead to an over-reliance on AI-generated content and reduce critical thinking, problem-solving, and creativity in the educational process?

Despite the fear or celebration, it is a fact that ChatGPT is changing the traditional learning landscape and forcing everyone to rethink and remodel education. As ChatGPT becomes more commonplace, it will be critical to calibrate expectations about its capabilities. One thing is clear: the use of AI in academia is a complex and nuanced issue that cannot be wished away. We cannot afford to bury our heads in the sand, for even there, the GPT storm will soon blow that sand away.

Accountability

Every institution of higher learning must be held accountable in ensuring that students are equipped with the knowledge and skills they need to

navigate this new technological terrain, while also upholding the ethos that keep higher education credible. We owe this to this generation and the next!

My intention here is not to give us a technical or academic treatise of this subject, but rather to kickstart an intentional dialogue toward this direction. I remember receiving my first COVID-19 vaccine, and then a few weeks later going for the “booster” vaccine, and breathing a false sigh of relief that the bad dream was finally over [story for another day!]. The efficacy rate of the GPT detectors will soon drop in the face of stronger mutations and variants of the “GPT virus.” Unlike with COVID, the new variants of GPT will only keep getting exponentially stronger. Should we develop a resistance to this new plague in our pedagogy and halls of academic exchange, or do we simply submit in defeat, embrace our AI overlords and let them colonize our skill to write, to create and to think critically for ourselves?