The scramble to catch AI-generated essays has masked a far more unsettling reality: higher education has been staring at a mirror, not a weapon. In a June 2026 essay, higher education leader John M. Fuchko III cut through the noise with an argument that has since forced campus administrators to rethink their entire posture toward generative AI. Rather than a tool to be adopted or banned, artificial intelligence, Fuchko claimed, is a force that tests what colleges actually believe about learning.

Fuchko’s assertion landed like a bombshell on a landscape already littered with contradictory policies. Universities had spent three years careening from outright prohibition to begrudging acceptance, often in the same academic term. Honor code tribunals bulged with cases. Faculty burnout spiked as instructors became amateur forensic linguists, hunting for the telltale flatness of machine prose. And yet, all the while, students were packing lecture halls, submitting assignments, and earning degrees—leaving everyone to wonder: what, exactly, were they learning?

The plagiarism panic was never just about cheating. It exposed an institutional anxiety about what a college education is supposed to produce. If a student can prompt an LLM to write a passable term paper in 20 seconds, then the transactional model of education—where knowledge is deposited and then withdrawn at exam time—collapses. Fuchko’s essay pushes past the detection arms race and asks a harder question: what if the real crisis is that we have designed learning environments where AI can outperform humans without even breaking a sweat?

The Detection Dead End

The initial institutional response to generative AI was almost entirely defensive. Turnitin launched an AI detection module in 2023, and within months, campuses were awash in false positives. Students, particularly non-native English speakers, found their original work flagged at alarming rates. Vanderbilt University famously disabled Turnitin’s AI detection tool after a student outcry. By 2025, a consensus had formed among academic integrity researchers that automated detectors were unreliable and, worse, inequitable.

But the retreat from automated detection left a policy vacuum. Many institutions fell back on suspicious-instructor-report workflows, which simply shifted the burden back onto faculty. The result was an exhausting cat-and-mouse game: students used AI to write, faculty used AI to detect, and the educational transaction became a battle between two black boxes. Lost in the process was any meaningful conversation about why students were turning to AI in the first place—and what that flight from effort signaled about the value of the learning itself.

Fuchko’s essay reframes the problem entirely. The issue is not that AI exists but that colleges have forgotten their own mission. If learning is merely the acquisition of demonstrable knowledge, then AI is a superior learner. But if learning is formative—if it is the slow, often painful process by which a person develops the capacity to think, to discern, to wrestle with complexity—then AI cannot replace it. The real task, he argues, is to build institutions that make that kind of formation inescapable.

Learning as Formation, Not Transaction

The concept of “formation” has deep roots in liberal education, stretching back to the medieval university and beyond. It’s the idea that education shapes character and intellect together, that the student who emerges after four years is not just better informed but fundamentally different in her habits of mind. That vision has been steadily eroded by an efficiency-obsessed culture that prizes measurable outcomes and speedy credentialing. AI merely accelerates that erosion.

Fuchko points to a stark example: if a student can complete an entire course by outsourcing the cognitive work to a language model, then the course was not asking her to do anything cognitively worth doing. The response cannot simply be to make assignments cheat-proof. It must be to design assignments that are cheat-proof because they are authentic—because they require the student to bring something of herself that no algorithm can simulate.

That shift is already underway in pockets of innovative pedagogy. Oral examinations, once relegated to graduate-level defenses, are making a comeback in undergraduate programs. Project-based learning, where students build portfolios that evolve over time, defeats the one-off prompt. Collaborative real-time writing, observed and iterated in class, restores the messy process of thinking to the center of the curriculum. These are not Luddite retreats but intentional designs that treat AI as irrelevant rather than adversarial.

The challenge, as Fuchko notes, is scale. Small seminars and studio courses can pivot to formative assessment with relative ease. A 300-student introductory lecture, on the other hand, was already a machine for processing learners. Adding AI to that equation has not created a new problem; it has exposed a preexisting one. The question becomes whether institutions are willing to radically restructure the economics of their offerings—fewer large lectures, more labor-intensive mentorships—to defend the heart of the educational project.

The Faculty Crucible

If learning is formation, then the faculty role cannot be reduced to content delivery and assessment. Fuchko’s essay underscores that professors must become something closer to intellectual coaches, walking alongside students as they struggle with material. That demands a level of time, trust, and relationship that mass production models of higher education have systematically dismantled.

Adjunctification and rising faculty workloads make this particularly painful. A contingent professor teaching five courses across three institutions cannot possibly offer the kind of formative attention that the AI moment requires. So while the op-ed calls on university leadership to articulate a vision, it implicitly indicts a business model that treats faculty labor as a cost to be minimized. Defending learning as formation may well require defending the faculty who are its only possible guardians.

There is also a generational divide. Younger scholars, many of whom grew up with digital tools, are often more comfortable incorporating AI into their teaching—sometimes as a collaborative brainstorming partner or a translation aid. Older faculty, steeped in a tradition of scholarly craft, may see any use of AI as a dilution of rigor. Fuchko resists the binary. He argues that the point is not to either embrace or shun AI but to integrate it in ways that serve, rather than substitute for, formative goals. The key test: does the tool help the student grow in capacities she will carry beyond the assignment?

Governance in the Gap

When the goal shifts from catching cheaters to designing for formation, institutional governance must change accordingly. Fuchko calls for a new alliance between academic affairs, student development, and information technology—three silos that rarely collaborate in meaningful ways. IT departments, accustomed to being order-takers for software platforms, need a seat at the educational design table. Student affairs professionals, who see students’ holistic development in ways that classroom instructors may not, become crucial partners.

Some campuses have begun experimenting with “AI in Learning” task forces that cut across these divisions. Their charge is not to produce a static policy document but to launch ongoing experiments: rethinking core curriculum requirements, piloting alternative credentialing, and building faculty development programs that teach the art of formative assessment. The goal is institutional learning, not just regulatory compliance.

That institutional learning requires humility. Fuchko’s essay closes with an uncomfortable observation: higher education has spent decades championing critical thinking as its stock-in-trade, yet its own response to AI has been remarkably uncritical. Campuses adopted detection tools without thorough validation, rushed to write policies that treated students as potential adversaries, and outsourced moral reasoning to software companies. If colleges cannot model the very deliberation they claim to teach, then students are right to be cynical.

The Road Ahead

The argument John M. Fuchko III advanced in June 2026 has not lost its urgency. If anything, the intervening months have sharpened its relevance as universities vet the next generation of AI tools, from AI teaching assistants to adaptive learning platforms. The lure of efficiency will remain powerful, especially in an era of enrollment cliffs and budget crises. But efficiency is not education. The institutions that will survive the AI upheaval are those that reclaim their foundational purpose: not to process students but to transform them.

That transformation depends on a bet that formation, not just information, still matters. It requires building academic communities where intellectual risk is safe, where failure is part of growth, and where AI is just another tool in the toolbox—powerful, but not precious. The hardest work lies not in updating honor codes but in rediscovering what it means to teach a human being how to think. And that, Fuchko insists, is a challenge no algorithm will ever solve.