Artificial Intelligence in Engineering Education: Opportunities, Risks, and a Human-Centred Framework for Responsible Integration

Abstract

Artificial intelligence (AI), and in particular generative AI built on large language models, is reshaping higher education and professional engineering practice. Engineering education is directly affected, because engineers are increasingly expected not only to use AI-enabled tools but also to understand their limitations, failure modes, and ethical implications. This paper presents a narrative review of AI in engineering education, focusing on teaching, learning, assessment, curriculum design, academic integrity, and employability. It does not report new survey or experimental data; instead, it synthesises recent international policy guidance and peer-reviewed scholarship to propose a responsible, human-centred framework for integrating AI into engineering education. The review indicates that AI can support personalised learning, programming assistance, simulation, design exploration, feedback generation, laboratory preparation, and administrative work. Uncritical adoption, however, may amplify risks related to hallucinated outputs, plagiarism, overdependence, skill erosion, algorithmic bias, privacy and data-security exposure, unequal access, and weakened assessment validity. The paper argues that engineering institutions should move from prohibition-based approaches toward transparent, outcome-based, and discipline-specific AI policies. A practical framework is proposed around five pillars: AI literacy, curriculum integration, assessment integrity, ethical governance, and faculty capacity building. The paper concludes that AI should not replace engineering fundamentals, mathematical reasoning, experimentation, design judgement, or teacher guidance; rather, AI should be treated as a powerful but fallible tool that strengthens engineering education when used with verification, transparency, and human responsibility.