2026 ASEE Annual Conference & Exposition

The ways that generative AI (e.g., tools built on and around Large Language Models) have been thrust upon both engineering students and educators has many of us confused and apprehensive about how the tools might impact critical issues like teaching, learning, and assessment. And while there are some ways AI might make positive contributions to engineering, there are also looming questions about academic integrity, the impact on traditional forms of assessment, and the role of higher education in society. This paper offers a framework grounded in diverse educational theories that can help engineering educators engage with AI in ways that discourage the use of AI tools in ways that subvert critical learning process. We focus on three areas: the individual, the relational, and the contextual. By attending to individual student motivations, educators can leverage student autonomy and control over learning to encourage productive AI practices that avoid extrinsic reward seeking and the disingenuous behaviors that result from it. In each area of focus, we offer both theoretical overviews and practical examples. By focusing on the relational aspects of learning, educators can create social, embodied learning experiences that discourage the use of AI to “farm out” important interactions, reflections, and the learning that accompanies authentic experiences. Third, when students can see how their learning has consequences and exists within a larger disciplinary context, the inherent value perceived in learning can encourage productive AI use through more critical engagement with disciplinary norms and methods of communication. Taken together, these three aspects can offer educators a way to design learning environments that avoid the ratcheting up of surveillance, an uncritical application of things like blue books, or any other aspects of what we see as an increasingly untenable game of cat-and-mouse when it comes to AI use in engineering courses.