AI Killed the Take-Home Essay. COVID Killed Attendance. Now What?

Reclaiming Learning in an Age of Distraction and Artificial Intelligence

In Spring 2026, I am going to do something I have never done in over 40 years of college teaching.

Every class session will begin with a 5-10 minute in-class writing prompt—handwritten, based on that day’s readings, collected and reviewed. Attendance will be mandatory and enforced. And students’ final grades will no longer depend primarily on essays written outside of class, where I cannot verify authorship or process.

Instead, assessment will center on activities I can actually observe: students will introduce class sessions, orchestrate discussions, deliver oral presentations, and field questions from their classmates. The work that matters will be the work I witness.

This represents a wholesale abandonment of principles I have defended throughout my career. I have always believed that college should cultivate intellectual independence, adulthood responsibility, and intrinsic motivation.

I resisted mandatory attendance because I wanted students to value learning for its own sake, not come to class out of fear of penalty.

I assigned traditional essays because I believed that serious writing required time, solitude, and the freedom to struggle privately with ideas.

I trusted students to do the reading and trusted that their essays reflected their thinking.

I no longer believe that the conditions which made these principles reasonable still exist. Something fundamental has broken. The pedagogical world I was trained for has dissolved with startling speed. And I find myself, late in my career, having to rebuild my practice from the ground up—not because I have lost faith in liberal education, but because I am trying desperately to preserve it.

What has forced this reckoning? A convergence of disruptions that have destabilized every assumption on which traditional college teaching depends.

The pandemic accelerated existing trends. Ubiquitous technology fractured students’ attention. Economic pressure instrumentalized learning. Therapeutic culture reshaped students’ relationship to intellectual challenge. And now artificial intelligence has rendered our primary tools for assessing learning—essays, exams, analytical writing of any kind produced outside our direct observation—functionally obsolete.

During this fall’s semester, I assigned my undergraduates weekly outside-of-class essays. Many are now interchangeable—uniformly well-organized, articulate, and confident, yet oddly generic and curiously detached from the specific issues we’re discussing in class. Most troubling, several papers made claims about sources we had never examined, complete with plausible-sounding analysis.

I could not definitively prove that AI had written these essays, but I could no longer assume that human minds had wrestled with the material. I could no longer trust the primary evidence on which I had always relied to understand what students were learning.

This semester, I realized I was teaching in a world I no longer understood, using methods designed for conditions that no longer obtained. The machinery of college teaching had broken, and I had been operating by sheer inertia, pretending it still worked.

Spring 2026 is my attempt to stop pretending. Here, I will do my best to explain why that pretense has become untenable, what forces have made it so, and what teaching must now become if genuine learning is to survive.

What Has Changed

Many students now regard class attendance as optional, shaped by remote learning during COVID, increasingly flexible high school environments where asynchronous work became normalized, and a general expectation that content will always be available elsewhere—recorded, transcribed, or summarizable by AI.

Reading loads and writing expectations have eroded, not because faculty abandoned standards, but because enforcing those standards became impossible when students arrived chronically under-prepared and overwhelmed.

The omnipresence of technology has fractured student focus in ways that are cognitively and pedagogically unavoidable. Multitasking is not a choice students make; it is the default cognitive mode in which they operate.

Meanwhile, a dramatic rise in official accommodations has reshaped the pace, structure, and assessment of almost every course. In my classes, over 25 percent of students now have documented accommodations—most commonly extended time on exams, but also alternative assessment formats and flexibility around deadlines.

These accommodations reflect genuine needs, and honoring them is both legally required and ethically essential. But when such a large percentage of students require structural modifications, something systemic has shifted.

The “standard” course design is no longer standard, yet most faculty receive no pedagogical training for designing courses that accommodate such variation without either diluting rigor or becoming logistically unmanageable.

The economic pressures reshaping student behavior deserve equal attention. Today’s undergraduates carry debt loads that previous generations would have found incomprehensible. Many work 20 to 30 hours per week—not for spending money but to cover rent and food.

They approach college instrumentally because they must: the stakes are too high, the costs too crushing, to treat education as an end in itself. When students skip class to work a shift, or submit perfunctory essays because they have no time for revision, or choose courses based on workload rather than interest, they are often making rational decisions within an irrational system.

Any serious reckoning with post-COVID teaching must acknowledge that students’ relationship to learning has been deformed by economic forces we did not create and cannot individually solve—but which we must pedagogically navigate.

And now there is AI—one of the largest disruptions to teaching and learning since the invention of print. A professor can no longer assume that an elegant essay reflects a student’s analytical ability, that a discussion post indicates the reading was completed, or that a problem set demonstrates mathematical understanding.

The traditional forms of evidence for learning—original writing, independent analysis, sustained engagement with complex texts—have been rendered unreliable.

But the challenge runs deeper than assessment. AI has created an epistemic crisis for students themselves. Why struggle to formulate an argument when ChatGPT produces one instantly? Why spend hours wrestling with a difficult text when an AI can summarize it in seconds?

Students increasingly doubt the value of their own cognitive labor. The machine thinks faster, writes more fluently, and never suffers the painful uncertainty that accompanies genuine intellectual work.

We are producing a generation that has access to infinite information and sophisticated analytical tools, but lacks confidence in their own ability to think—and increasingly lacks practice in the sustained attention that deep thinking requires.

What, exactly, do we mean anymore by “a student’s own work”? The phrase assumes a clear boundary between autonomous and assisted thinking, but that boundary has become philosophically unstable.

If a student uses AI to generate an outline, is that assistance or outsourcing? If they feed their draft to AI for revision suggestions and implement them, whose prose is it? If they use AI to summarize readings before class discussion, have they “done the reading”?

These are not hypothetical questions. They describe the daily reality of undergraduate intellectual work in 2025, and we have not yet developed coherent answers.

The result is not that students are less capable or less motivated than previous generations, but that the instructional ecosystem has collapsed while a new one has not yet emerged. We are teaching in the gap between two worlds.

What Must Change

The future of effective teaching requires rethinking some long-standing habits:

Attendance now matters because learning is no longer anchored by a shared cultural expectation of presence. Students need to prioritize class participation and make showing up an essential part of their overscheduled lives. Mandatory attendance is not punitive; it is a form of pedagogical care that acknowledges the competition for students’ attention and asserts that this time, this space, this community matters.

In-class writing matters because it is one of the few remaining reliable ways to gauge a student’s own thinking. I will begin each class session with a 5-10 minute analytical response to a question about the reading—handwritten, ungraded for content but checked for completion, discussed immediately afterward.

The writing will often be rough, frequently confused, sometimes superficial. But it is theirs. I can see who has engaged the text, who is struggling with it, who has not yet opened it. More importantly, students can see their own thinking emerge on the page. They discover that they have ideas—partial, uncertain, underdeveloped, but genuine. This discovery is increasingly rare and increasingly precious.

Active responsibility and public performance matter because they shift learning from private production to observable demonstration. I will no longer ask students to produce written work I cannot witness.

Instead, each student will be responsible for introducing a class session—presenting the core arguments of that day’s readings, identifying key tensions or problems, proposing questions for discussion.

They will take turns facilitating discussions, which means not merely participating but actively orchestrating the conversation: drawing out quiet voices, pressing superficial claims, connecting contributions, and synthesizing insights.

And they will deliver oral presentations on research topics of their choosing, then defend their findings against questions from classmates and from me.

This approach inverts the traditional relationship between student and instructor. I am no longer the sole authority responsible for making the class work. Students become co-creators of the learning environment, accountable not just to me but to each other.

The student introducing Tuesday’s session on the Federalist Papers cannot simply summarize what an AI told them about Madison’s argument. They must demonstrate live, in front of their peers, that they have read carefully enough to guide others through the text. The student facilitating discussion cannot hide behind polished prose. They must think on their feet, respond to unexpected turns in conversation, manage disagreement, and synthesize complex exchanges in real time.

The student presenting research findings cannot submit a paper and disappear. They must stand before the class and articulate their claims clearly enough to withstand scrutiny, then revise their thinking publicly when challenged.

Can students prepare for these performances using AI? Of course. They can use it to help organize their thoughts, generate potential discussion questions, or outline their presentations. But the performance itself cannot be outsourced.

When a student stands in front of the class struggling to explain a passage they have not actually read, or facilitating a discussion about ideas they do not genuinely understand, or defending claims they cannot support—the gap between AI-generated preparation and authentic comprehension becomes immediately, painfully obvious. More importantly, it becomes visible to them. They discover in real time what they know and what they have merely simulated knowing.

Learning becomes visible not through submitted drafts but through embodied performance.

What students gain is a different kind of confidence—not the confidence of having produced a polished paper, which may or may not reflect their actual understanding, but the confidence of having demonstrated competence publicly and repeatedly.

They learn that they can read a difficult text well enough to teach it to others. They discover that they can manage intellectual disagreement without collapsing into relativism or hostility. They find that they can articulate complex ideas under pressure and revise their thinking when challenged.

These are the capacities that the traditional essay was supposed to cultivate but increasingly fails to measure. Public performance makes them unavoidable.

There is also a profound shift in classroom culture. When students know that each of them will be responsible for making the class work—not occasionally but regularly—they begin attending to the quality of discussion differently.

They arrive prepared because they know their peers are counting on them. They listen more carefully because they may be the one responsible for synthesizing what they hear. They take intellectual risks because the environment is one of shared responsibility rather than individual performance before an authority figure. The classroom becomes, once again, a genuine intellectual community rather than a theater where students perform for a grade.

This approach is demanding—for students and for me. It requires that I teach students how to introduce readings effectively, how to facilitate discussion without dominating it, how to construct and deliver presentations that invite genuine engagement. It means that class sessions will sometimes falter when student facilitators struggle, that discussions will sometimes meander or stall, that presentations will sometimes be rough or incomplete.

I am trading the controlled efficiency of instructor-led classes for the messier, more unpredictable process of student-driven learning. But that trade feels necessary. The controlled efficiency was producing an illusion of learning. The messiness may actually produce the real thing.

The Costs and Complications

None of this is cost-free. This approach requires significantly more instructor time and energy than the lecture-and-exam model or the assign-essay-and-grade approach. In an era of adjunctification, these recommendations may seem like cruel fantasies. The structural exploitation of academic labor is real, consequential, and worsening. But acknowledging that reality does not absolve those of us with the privilege of tenure-track positions from rethinking our practice.

The pedagogical approach I am advocating works most naturally in small to mid-sized classes—seminars and discussion courses of fifteen to forty students. Larger classes require other solutions: frequent low-stakes quizzes, strategic use of teaching assistants for small discussion sections, creative deployment of technology for real-time feedback. But I will not pretend the translation is simple.

One more complication: Is this a temporary disruption or a permanent transformation? Are we seeing a COVID cohort effect that will eventually normalize, or a generational shift in how young people approach learning?

My instinct, based on limited evidence and much uncertainty, is that we are witnessing something more lasting. The forces that destabilized traditional teaching—ubiquitous technology, economic precarity, the erosion of shared reading culture, and now AI—are not temporary.

They are structural features of contemporary life. We should design our teaching for the world that exists, not the world we wish existed.

The Democratic Stakes

The challenge we face is not merely pedagogical. It is political in the deepest sense.

Liberal democracy depends on citizens capable of critical reading, independent analysis, and sustained engagement with complexity. It requires people who can distinguish evidence from assertion, who can follow an argument across multiple pages, who can revise their thinking when confronted with new information, who can tolerate ambiguity and uncertainty without retreating into simplistic certainty.

These are not innate capacities. They are learned practices, cultivated through education.

If universities cannot reliably foster these capacities—if we cannot distinguish authentic from artificial thinking, if we cannot create conditions where students practice the slow, difficult, frustrating work of making sense of the world on their own terms—then we are failing at our most essential social function.

The consequences extend beyond individual students. A society in which citizens outsource thinking to algorithms, in which reading atrophies from lack of practice, in which the capacity for sustained attention becomes rare—such a society cannot maintain democratic institutions. It becomes vulnerable to manipulation, demagoguery, and the gradual hollowing out of public discourse.

This is why the current moment is not merely a pedagogical puzzle to be solved with better assignments and classroom management. It is a crisis of civic capacity, playing out in thousands of classrooms, largely invisible to the public, inadequately addressed by institutions, and profoundly consequential for the future of democratic life.

Building What Must Be Built

The world in which we once taught is gone. Those of us old enough to remember it may mourn its passing, but nostalgia is not a pedagogical strategy. The challenge, and the opportunity, is to build the world in which genuine learning can still flourish.

This requires courage—the courage to abandon practices we cherished, to risk seeming authoritarian when we impose structure, to acknowledge that our previous ideals depended on conditions we did not create and cannot restore.

It requires humility—the recognition that we are improvising, that we do not yet have definitive answers, that some of our interventions will fail and require revision.

It requires solidarity—the willingness to share what works, to learn from colleagues across institutions and disciplines.

Most of all, it requires clarity about what we are trying to preserve. We are not defending a particular pedagogical style or a nostalgic vision of college life. We are defending the possibility of authentic intellectual development—the cultivation of human capacities for careful reading, sustained thinking, precise expression, and ethical engagement with ideas. These capacities are essential to human flourishing and democratic survival.

The post-COVID university has changed irrevocably. Our responsibility is not to resist that change but to shape it—not by lowering standards, but by rebuilding the structures through which students can meet them.

The task before us is to construct a teaching environment that emphasizes engagement over passive consumption, authenticity over polish, and process over product. If we want students who read carefully, write seriously, and engage ethically with ideas, we must design classrooms where those practices are unavoidable—and where the rewards of those practices are tangible, human, and real.

We must build, in other words, conditions for genuine learning in an age designed to make such learning nearly impossible. It is urgent work. It is necessary work. And it is work we cannot postpone.