Critical Evaluation of the Potential of Large Language Models in Bioscience Higher Education: A Conversation with ChatGPT

Abstract

The possibilities for integrating generative artificial intelligence (AI) and large language models (LLMs) into higher education may revolutionise approaches to pedagogical practices and  curriculum design, while LLMs could be transformative in how students approach their learning. This conversation with ChatGPT, and associated critical evaluation, provides an insight into the capabilities and limitations of LLMs and informs on the possibilities for incorporating AI into bioscience education, teaching modalities, assessment and feedback practices and curriculum design, with a particular focus on  bioscience education. The conversation highlights how LLMs can facilitate personalized feedback, tutoring, and concise explanations of scientific terms, enhancing student comprehension, self-directed learning, and critical thinking. However, there are concerns regarding the risks of LLMs, such as plagiarism and breaches of academic integrity, algorithmic bias and limitations in contextual understanding. Despite these limitations,  AI offers opportunities for enriching undergraduate bioscience curricula by integrating innovative teaching strategies and assessment modalities aligned with subject benchmark statements. Future research directions include exploring ethical implications, equitable access and digital literacy training in higher education settings. Overall, the integration of LLMs in bioscience education offers significant potential for innovative pedagogical approaches and transformative learning experiences.