Generative AI Tools in Higher Education: Benefits, Limitations, and Ethical Risks

Abstract

In recent years, neural networks have experienced rapid development and are increasingly implemented across various areas of human activity, including education, which necessitates research into their effectiveness for didactic purposes. This article investigates the opportunities and challenges associated with using generative AI in higher education, particularly for professional English language training among natural science students. The study employs theoretical general-logical methods (analysis, synthesis, generalization, induction, deduction) and empirical methods (pedagogical observation, experiment, description, survey). Its theoretical and methodological foundation is based on cognitive psychology, cognitive pedagogy, and AI theory. The article critically analyzes the literature, highlights the advantages and limitations of implementing generative AI in educational processes, and examines the didactic potential of neural networks such as ChatGPT, Copilot, and AI Tutor. A pedagogical experiment was conducted, exploring AI’s roles as teacher, editor, and foreign-language interlocutor. The study identifies key pedagogical capabilities of ChatGPT in fostering students’ cognitive and linguistic development, including critical thinking, creativity, cognitive interest, motivation, reflexivity, and language skill acquisition. Results demonstrate AI’s effectiveness in compiling thematic glossaries, contextualizing terminology, testing vocabulary, and conducting dialogues, which enhance cognitive engagement. Final assessments revealed that students in the experimental group achieved higher proficiency in geological English terminology, scientific text comprehension, and listening skills.

Ähnliche Arbeiten