Assessing the Trustworthiness of Generative AI Used in Higher Education

Abstract

In the rapidly evolving area of artificial intelligence (AI), we find ourselves at a path-breaking juncture in a global technical paradigm. Remarkable advancements have propelled large language models (LLMs), such as ChatGPT (3.5/4), Bard, and Claude, to a level where their advanced text-generation capabilities mimic human-like fluency and the ability to refine and transform text to such an extent that discerning their output from human-authored content becomes a formidable challenge. The foreseeable proliferation of such advanced models and the continuous evolution of their capabilities necessitate their inclusion in the landscape of higher education and research. Therefore, it is imperative to recognise their presence and profound implications, prompting a deliberate exploration of their integration within the academic sphere. It is important to view the emergence of LLMs as an opportunity to enhance the education domain. Degree programs and educators are encouraged to incorporate AI into their curriculum to prepare students for a future society where AI-based applications will be extensively utilised. It is crucial to use AI-powered text-generation tools in a controlled and guided manner considering the potential ambiguity and reliability issues that might arise. In the educational sector, their usage should be limited to situations where they would aid in promoting student learning. To take an example, at the EU level, an AI regulation is under preparation, which will also apply to AI systems in education. In addition, there is an ethical policy on AI and its use, as well as an ethical code for teachers. The guidelines from various academic institutions may be further specified considering future regulations and technological developments.

Ähnliche Arbeiten