Opinion | Stop Pretending You Know How to Teach AI

Colleges are rushing to declare students “fluent” in new skills. But one major issue remains: nobody fully understands what that fluency entails.

This essay is excerpted from a special report titled Leading in the AI Era, available from the Chronicle Store.

AI Fluency at Ohio State University

Earlier this year, Ohio State University launched an AI Fluency program promising that, starting with the Class of 2029, every graduate will be “fluent in their field of study, and fluent in the application of AI in that field.” Faculty across departments have been encouraged to find ways to integrate generative AI into their disciplines.

Reflecting on this, I considered a classical question: How might AI help students—or university administrators—explore Socrates’ paradox from Plato’s Apology?

“I neither know nor think I know.”

The Challenge of Defining AI Literacy

Over the last two decades, education reform has followed a pattern we can call “technology literacies.” When new technologies emerge—such as tablets, smartphones, the web, social media, programming languages, or AI—experts define related skills as a “literacy” or “fluency.” Policymakers then urge schools to incorporate these skills into their curricula.

However, this cycle often happens without clear consensus on what true fluency means or how best to teach it.

Access and Subscriptions

To read the full report, one can subscribe starting at $10 per month for unlimited access. A free account offers limited articles each month, along with newsletters, job postings, salary data, and store discounts.

Summary

The rush to teach AI fluency in higher education highlights a critical gap: we lack a shared understanding of what AI fluency truly means or how to effectively cultivate it.

The Chronicle of Higher Education — 2025-11-06