Abstract
With the advent of Artificial Intelligence (AI), many institutions of higher education have devolved both creativity and accountability for AI policy to the level of the individual instructor. At the 2025 Conference on College Composition and Communication, Harry Denny and Brent Cameron reviewed the decision-making process among Liberal Arts & Sciences Deans in the Big 10 conference about AI. They reported the general consensus among these deans was to devolve policy decisions to the level of individual instructors (Denny & Cameron, 2025). Anecdotally, this consensus to place responsibility at the individual appears common across higher education – Denny and Cameron note similarities between Big 10 schools and West Coast institutions. Devolution to the instructor’s level certainly provides flexibility and prevents institutional over-reach. Yet Denny and Cameron describe a “paradox or contradiction” wherein a university-level problem has no university-level fix. Denny noted specifically that universities rushed to “brand/ish” AI to promote their institutions without concomitant policy for its use; Cameron noted the potential for iniquities without consistent policy. Emily Hall provides a snapshot that captures some of the conflicting demands on instructors in this situation: “frustrated pleas for technology to detect AI ‘cheating’ arrive alongside invitations to AI implementation workshops featuring faculty integrating gen AI into writing assignments and activities” (Hall, 2025). In a phrase: individual instructors create, articulate, and adjudicate what their institutions abdicate. The upshot of this situation is that individual instructors who can find cunning adaptations are likely to lead from below. This paper will provide an account of one case that inverted several aspects of this dynamic. In two sections of a second-level communication class, students were challenged to author AI policies for a mid-sized firm, and at the same time students were required to use large language models (LLMs) to write those policies. The upshot of this is a general form of assignment design that transforms the use of LLMs from a shortcut to the royal road to rhetorical awareness.
Rights
© 2026 The Author
Recommended Citation
Thomas, Evan, "Speech-Acts and Student-Led LLM Experimentation" (2026). The Interdisciplinary Journal for Research on Pedagogy and Learning Design, Vol. 1, Issue 1., Fall 2026. 19.
https://openprairie.sdstate.edu/ijrpld/19
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License