2026 ASEE Annual Conference & Exposition

This research paper investigates patterns in how engineering instructors use metaphors to conceptualize artificial intelligence (AI), examining how they shape teaching practices and student understanding during a time of significant technological change in higher education. At least since Socrates, many have argued that metaphors structure the ways we conceptualize the world, relating and reasoning about one thing in terms of another. Conceptual metaphor theory suggests that figurative language structures concept understanding; central to learning by teaching us what things “mean.” Metaphors allow individual and social experiences to be abstracted, recognized as entities, and related to others so as to establish “meaning.” Accordingly, engineering education is steeped in metaphor, texturing the ways we describe, teach, and construct engineering theories and applications.

Many metaphors have percolated the public consciousness with the widespread use of AI. AI as a black-box: an inscrutable, physical thing with incomprehensible complexity. AI as an imperfect copier: stochastic prediction algorithmic models that parrot plausible-sounding text without understanding its underlying meaning. Or, AI as simulators: a technology existing across a “multiverse” of roles simultaneously, being fine-tuned with each prompt. These examples illustrate some ways AI is conceptualized; however, current AI metaphor analyses are not always focused on educational settings. Rather, metaphors are consolidated from wide-ranging materials, including academic research, law and legislative materials, social media discourse, science fiction, and periodicals. Literature focused on AI metaphors in education largely concern those found among students, such as those at the K-12 or undergraduate levels. Consequently, less is known about the AI metaphors held and espoused by instructors: those who may influence the AI metaphors observed among students and inform how students, in turn, think about AI.

As part of a large-scale study of engineering faculty members across the United States about GAI in STEM higher education, this preliminary analysis investigates patterns in instructors’ metaphor use to describe AI. During one-on-one interviews, 57 STEM higher education instructors were asked to describe a metaphor they use to conceptualize AI. Results reveal metaphors about the nature of AI, how AI operates, how people and society relate to AI, and AI’s possible impacts. Their usage suggests differences in instructors’ mental models of AI, varying across these dimensions in emphasis and description. For example, some relate AI to calculator-like tools, while others show limitations of this analogy by stressing that AI, unlike calculators, do more than merely provide calculations for students. These perspectives extend previous literature investigating AI metaphors, underscoring that instructors are vital in helping students understand their learning. They give words to students’ feelings, ideas, questions, and concepts that facilitate student development. Understanding instructors’ AI metaphors is therefore critical to future educational policy and decision-making, because no single conceptualization has ever, or will ever, describe “AI.” Yet, these metaphors offer insights into how stakeholders conceptualize this set of technologies during a period of change in higher education.

The authors prefer to present this work as a mini-demonstration, engaging attendees in exploring their own AI figurative language and the implications for faculty development practice.