AI Will Win if We Let It

What appears to be a learning crisis (declining test scores, unwillingness or inability to read what teachers assign, resistance to challenge, etc.) might be more fruitfully addressed as a a crisis in how teachers relate to learners.

The educational crisis evidenced in the media under the banner of NAEP is a symptom of something much larger the citizenry is avoiding looking at directly—a loss of trust in young people, a pervasive sense of chaos, a hardening stance toward teacher control and learner compliance. Into this season of discontent with public schools comes AI.

A social experiment is unfolding to change the status quo of schooling. Consider the emerging alternative model represented by Alpha School, a private PreK-12 network that promises to complete traditional academic instruction in merely two hours daily through AI-powered adaptive learning platforms.

Alpha employs what it terms “mastery-based progression,” wherein students work individually on third-party applications like Synthesis Tutor and Math Academy, advancing only after demonstrating complete comprehension of each concept through principles of spaced repetition and personalized pacing.

The model eliminates credentialed teachers in favor of less expensive and less well educated “guides” who provide motivational and emotional support rather than academic instruction, their roles transformed from content delivery to simply being there.

Students spend mornings engaged with AI tutors that generate individualized learning pathways, then devote afternoons to workshops focused on twenty-four identified “life skills” ranging from public speaking to entrepreneurship and financial literacy.

The system incorporates gamification through an internal currency called “Alpha Bucks” that students earn for academic progress and can deploy for passion projects or investments, explicitly motivating children by “giving them the gift of time” to pursue interests beyond standardized academics.

With tuition ranging from forty thousand to seventy-five thousand dollars annually depending on location, Alpha represents a market-based solution to educational inefficiency, acknowledging the comprehensive failure of mass-production traditions in schooling while simultaneously restricting access to families possessing substantial economic resources.

Alternatively, nonprofit Khan Academy has developed Khanmigo, an AI-powered tutoring assistant built on OpenAI’s GPT technology and designed explicitly to avoid simply providing answers to student queries. Khanmigo claims to operates according to Socratic pedagogical principles, refusing to complete homework or supply direct solutions, instead posing guiding questions that encourage students to arrive at conclusions through their own reasoning processes.

The regular school system integrates with Khan Academy’s extensive content library spanning mathematics, humanities, coding, and social studies, offering personalized tutoring tied to individual learning histories and progress patterns across more than one hundred thousand lessons, videos, and practice exercises.

For teachers, Khanmigo functions as an administrative assistant capable of generating standards-aligned lesson plans, crafting rubrics and exit tickets, and providing on-demand summaries of recent student work that illuminate patterns and identify areas requiring additional support.

The tool remains free for all teachers throughout the United States while school districts must pay fifteen dollars per student annually to cover computational expenses, and individual families can purchase access for four dollars monthly.

Unlike Alpha School’s complete replacement of credentials, Khanmigo positions itself as a supplement to existing classroom structures, providing the personalized attention of private tutoring at dramatically reduced cost while maintaining human teachers in their traditional roles.

Although Khanmigo takes the opposite stance vis a vis Alpha School in terms of providing the full package without credentialed teachers, it fails to radically change the status quo but supplements an already failing system.

There are good reasons to be concerned about autonomy and the personal freedom in childhood to enjoy peers in this emerging AI ecosystem. Learning among peers with a teacher opens free spaces for playful interpersonal experiences which isolation behind a screen forecloses.

Peter Gray, a research professor at Boston College, argues on Substack that “children today are less free than children have ever been in the history of humanity (with the exception of times of high child labor and slavery).” Adding AI in isolation intensifies this problem it seems to me.

Gray goes further, stating that it’s fair to say “our society is psychologically abusing children right now because of the way we restrict their freedom.” Gray is an expert in the significance of play in childhood as a substrate of robust growth and development

The panic associated with AI in education, however, has nothing to do with how children feel about learning. Indeed, many teachers already aren’t losing sleep over the fact that children are often bored. The big problem is student cheating..

With or without AI in the picture too many educators have no idea if and what middle snd high school students are actually thinking, reading, and writing unless it has to do with a targeted learning objective, and they may never have had the ability and the time to look deeply into student thinking way before AI.

How could high school teachers ever have deeply known about individual student’s literacy in their overcrowded classrooms? The reality is that students are always thinking, speaking, reading—just not thinking the thoughts teachers want them to think. These messy, real thoughts are what cognitive theorists call “extraneous,” a waste of brain power.

When we deny the role of joy and playfulness while learning, when the only thoughts from students we care about are those directly linked to the curriculum, we are diminishing freedom to learn. What world are we asking students to care about? And why should they?

Hannah Arendt writing in 1954 published a famous piece titled “The Crisis in Education” about the transformation of American high schools in the 1940s with great relevance to the question of authenticity, freedom to think, and freedom from intellectual domination.

In this composition Arendt responded to a changing global picture for education following the residual effects of death camps in Germany and nuclear bombs in Hiroshima. Arendt posed the question: What kind of world are we preparing children for? Why should they want to live in it?

Beginning in 1945, from a resolution at a vocational educators’ conference, the Life Adjustment Movement represented a radical shift in curriculum philosophy, one Arendt warned her contemporaries to resist.

The U.S. Office of Education created the Commission on Life Adjustment in 1947 to redefine progressive education. This reform was based on the assumption that 60 percent of high school students were neither college-bound nor suited for technical training, so they needed “life skills” education instead.

Students would study applied business math rather than advanced mathematics, and communication skills would replace traditional literature and grammar. The movement emphasized relevance to the needs of society, instrumentalism, social order, and patriotism as core principles, with particular focus on instilling Cold War-era patriotism to protect national security.

High schools shifted from institutions that “prepared young people academically and vocationally for the adult world” to more “custodial” institutions devoted to social management and conformity. Harvard president James B. Conant argued that parents should accept educators’ judgments about their children’s intellectual limitations, while the National Education Association’s Department of Secondary School Principals declared that traditional courses should be replaced with “fundamental categories of genuine student experiences,” the precursor to wood shop, home economics, and auto mechanics.

This reform kept more students in school during a period when dropout rates remained extraordinarily high. Trying to determine the dropout rate in U.S. schools is a notorious rabbit hole involving major statistical analyses applying proxy variables, but there is broad agreement that it remained a problem through most of the 20th century with a mysterious drop in the 1970s.

Progressive reformers faced a genuine democratic challenge during the 1950s: how to educate everyone, not merely an elite subset of the population. Europe sorted children at age eleven into academic and vocational trajectories.

America rejected this stratification as incompatible with democratic principles. Consequently, reformers attempted something radical: the comprehensive high school for all adolescents spending their days in peer-governed classroom environments.

The reforms failed and continue to fail, Arendt argued, though not because they were excessively progressive in orientation, granting greater presence to the learner. They failed because they subjected children to something considerably more oppressive than adult authority. “The authority of a group, even a child group, is always considerably stronger and more tyrannical than the severest authority of an individual person can ever be,” she observed.

An individual child becomes “a minority of one confronted by the absolute majority of all the others.” The inevitable outcome? “The reaction of the children to this pressure tends to be either conformism or juvenile delinquency, and is frequently a mixture of both.”

Educators imagined they were emancipating students by eliminating adult authority in the peer space. Instead they constructed what Arendt termed “the tyranny of the majority,” wherein every child must perform ceaselessly for their peers under conditions of relentless peer surveillance. Social media platforms did not invent this dynamic. They merely rendered visible what the peer-governed classroom had already demanded as the price of belonging.

The second catastrophic transformation involved reconceptualizing the nature of teaching authority itself. What Arendt identified was not primarily a deficit in teacher knowledge but rather a philosophical reorientation about what teachers knowledge teachers are permitted to claim.

Contemporary teacher preparation programs required aspirants to demonstrate pedagogical methodologies while systematically discouraging them from presenting themselves to learners as authentic adults possessing coherent worldviews worthy of transmission. Instead, they delivered standard curriculum content.

The credential to teach demanded a peculiar form of self-abnegation: teachers must position themselves as neutral facilitators of learning processes rather than as representatives of the adult world with legitimate perspectives about what matters and why.

The peer model, with its rhetorical commitment to equity and fairness, necessitates that teachers remain scrupulously blind to individual distinctions among students. They must treat each student with identical procedural neutrality, applying standardized methodologies without regard to particular capacities or needs.

This obligation extends beyond simple equality of treatment into a more profound requirement: teachers must not privilege their own judgment about the relative merit of student contributions, the significance of particular knowledge domains, or the hierarchy of intellectual virtues worth cultivating. Teachers must privilege evidence from work products using similar criteria.

The modern system cannot function without some mechanism of hierarchical evaluation and social control. Having stripped teachers of legitimate authority grounded in their unique relationship to knowledge and the adult world, the institutional apparatus provides them instead with raw coercive power: the capacity to assign grades according to their own lights. This substitution of power for legitimate authority represents perhaps the most corrosive corruption embedded within the peer-model paradigm.

Grading exercises dominion over children in ways that parental authority cannot and should not replicate. Parents possess natural authority derived from their role as protectors and guides into the world. Teachers in traditional models possess delegated authority derived from their designated mastery of curriculum and their position as trusted representatives of the broader adult community.

But the credentialed facilitator working within peer-model classrooms wielded something quite different: bureaucratic power to attach alphanumeric symbols that determine students’ future access to opportunity.

The teacher told to efface personal conviction, forbidden from asserting that some knowledge matters more than other knowledge, prohibited from presenting a coherent worldview different from the standard view nevertheless retained absolute discretion to evaluate student performance and to claim objectivity while encoding highly contestable judgments. We learned from research in the 1960s, for example, that ten different English teachers would score the same essay from an A to a F. This created a profoundly corrupting dynamic.

Students learned to perform for grades rather than to think for themselves. Teachers learn to wield evaluative power as a classroom management tool rather than as pure assessment of intellectual growth. The grade became currency in a system of exchange where everyone understands the transaction but no one understands the objective underlying value.

The teacher cannot say “I have lived longer, studied more deeply, and developed judgment worth respecting about what constitutes excellence in this domain.” Such claims violate the facilitator’s professional obligations to remain value-neutral, evidence-centric.

Yet this same teacher must assign differential grades that dramatically affect students’ life trajectories. On what legitimate basis? If the teacher can claim no authority grounded in superior knowledge or judgment, then grading becomes merely the exercise of arbitrary power, power students rightly experience as quixotic precisely because it lacks authoritative foundation.

This explains much of the intensity of grade negotiation, the constant appeals to fairness defined as mechanical equality, the transformation of education into transactional exchange. Students have correctly perceived that their teachers wield power without authority.

They respond by treating grades as commodities to be bargained for rather than as honest assessments rendered by authorities whose judgment they trust. The teacher, forbidden to claim authority but required to exercise power, becomes something worse than authoritarian. The teacher becomes a petty bureaucrat administering a system neither party deeply believes serves genuine educational purposes.

Arendt argued that “the educators here stand in relation to the young as representatives of a world for which they must assume responsibility although they themselves did not make it.” The teacher’s legitimate authority derives precisely from their willingness to say to the child, “This is our world,” and to stand behind that declaration with the full weight of adult conviction about what matters within that world.

But the credentialed teacher working within the peer-model paradigm is systematically prohibited from making such declarations while simultaneously required to exercise coercive power over students’ futures through grading.

Here the artificial intelligence crisis intersects decisively with this corrupted authority structure. Contemporary American education operates according to an industrial mass-production paradigm.

Thirty students occupy each classroom space. Standardized curricula dictate instructional content across institutional boundaries. Assessment mechanisms rely upon artifacts that furnish documentary evidence of completed labor and justify the assignment of grades.

The entire apparatus functions on the foundational assumption that students will generate quantifiable outputs: multiple choice tests, completed activities, analytical essays, mathematical problem sets, laboratory reports documenting experimental procedures, etc.

Then, suddenly, in November, 2022, a technology corporation provided assembly-line learners with sophisticated automation technology.

Within days traditional high schools screamed foul. Nobody asked for this. Nobody wanted to disrupt the peer model wherein everyone wanted to know what grade their friend got on an assignment.

Interestingly, ChatGPT posed no substantive threat to institutions like Phillips Exeter Academy. The first private secondary school in the country founded in the late 16th century, Exeter is the gold standard of elite schooling.

There, classrooms are constituted of individuals. Each handpicked learner is viewed as a highly gifted and trusted peer and learner. Twelve students convene daily around oval tables, Harkness tables, alongside an instructor hired precisely for possessing unique, profound mastery of the subject matter under consideration.

Students arrive to seminar intellectually prepared to lead and engage in challenging discussions after formulating compelling questions about complex topics and to channel collective inquiry. Crucially, the teacher at the Harkness table is expected to function as an authentic intellectual authority with a unique perspective on the world, someone whose deeper knowledge and more refined judgment commands respect without requiring coercion through grades.

The valued educational product in the Harkness method is not a polished essay submitted for standardized scoring but rather the caliber of reasoning demonstrated through sustained dialogical engagement in a variety of discourse practices.

Such authentic intellectual exchange resists algorithmic simulation. Artificial intelligence offers nothing of consequence to a genuine Socratic seminar mentored by someone who genuinely knows the territory—not the curriculum—and whose authority students recognize as legitimate.

Within factory-model classrooms, however, generative AI proves utterly devastating. Not because students engage in academic dishonesty, but because the technology exposes what the institutional apparatus actually valued all along.

Mass schools do not prize authentic, highly idiosyncratic critical thinking. They prize the efficient production of textual artifacts that can be converted into grades according to standard criteria to maintain institutional order and fulfill state-mandated actions.

They do not value substantive discussion guided by legitimate adult authority. They value compliance with assigned tasks that justify the exercise of evaluative power by credentialed facilitators forbidden from claiming genuine authority. Educators constructed an educational assembly line and christened it democratic pedagogy.

In November, 2022, students simply automated processes that were already fundamentally mechanical in nature. Their response cannot be so easily dismissed as ethically deficient.

They behaved rationally given the incentive structures adults deliberately constructed and maintained. More significantly, they responded to teachers who wield grading power without exercising their legitimate epistemological authority as committed disciplinarians, teachers who must maintain studied neutrality about what constitutes creditable learning while assigning grades that determine futures.

If teachers cannot present themselves as unique and capable representatives of the adult world, if they must exercise power through following a curriculum and passing out grades while being forbidden from claiming authority through knowledge and judgment, then students reasonably conclude that the entire system constitutes an arbitrary obstacle course rather than an initiation into something valuable.

The grade can rationally become a credential to be obtained through whatever means prove most efficient, including algorithmic generation of acceptable textual outputs.

The implicit bargain becomes “comply with my requirements, and you shall receive acceptable grades. Resist, and suffer evaluative consequences.” This transforms the teacher-student relationship from one potentially grounded in legitimate authority and mutual respect into one structured by coercion and strategic calculation.

Parents cannot replicate this corrupting dynamic because parental authority, whatever its limitations, rests on fundamentally different foundations. Parents need not justify their concern for children’s welfare through the need for bureaucratic credentials.

They need not pretend value-neutrality about what matters. They exercise authority naturally embedded in the relationship itself. But teachers in peer-model classrooms possess no such natural authority. They have only institutional power, power that corrupts precisely because it lacks legitimate authoritative foundation in the eyes of learners.

Arendt foresaw precisely this trajectory of institutional failure. She characterized it as the comprehensive bankruptcy of progressive educational philosophy. The reformers genuinely attempted to resolve the dilemma of universal education by eliminating capricious hierarchical adult authority, emphasizing kinetic activity over systematic learning, and establishing peer-governed social spaces.

Yet they never adequately addressed the foundational question: how does a democratic society provide opportunities for authentic intellectual formation to every citizen when it systematically prevents teachers from exercising legitimate intellectual authority while simultaneously requiring them to wield coercive, standardizing power through grading?

The comprehensive high school was conceived as democracy’s pedagogical answer to aristocratic selectivity. It rejected the European model, cutting off routes to higher order learning for lesser children at 11 years old.

Instead it devolved into an adolescent containment facility where credentialed facilitators, stripped of legitimate authority but armed with grading power, manage crowds through evaluation systems neither party believes serve genuine educational purposes.

Unable to devise methods for teaching thirty students with genuine effectiveness, administrators systematically reduced academic standards while maintaining elaborate grading apparatuses that create the appearance of differentiation and accountability.

Now the voucher movement threatens to abandon even the rhetorical commitment to democratic education as a shared public good. Its proponents advocate reconstituting quality schooling as a scarce market commodity accessible exclusively to households with adequate purchasing power.

Everyone else receives the discredited factory model, now thoroughly exposed as pedagogically bankrupt by the very technological instruments upon which it depended, staffed by teachers professionally prohibited from claiming the authority that might render education interesting and meaningful yet required to wield the grading power that makes education coercive.

This constitutes the crisis Arendt identified with such devastating clarity seven decades ago. Contemporary educators remain ensnared within identical contradictions.

Returning to European-style academic sorting at age eleven remains politically untenable. American civic culture will not countenance such explicit stratification.

Yet the mass-production educational model has manifestly failed to deliver on its democratic promises, particularly after systematically replacing legitimate teacher authority with bureaucratic grading power. And now artificial intelligence has rendered that failure undeniable to anyone willing to examine the evidence.

The salient question is not how regulatory authorities might prohibit ChatGPT usage. The question is whether American society possesses sufficient political will to restructure schools around genuine learning or whether it will capitulate to the voucher movement’s vision of education as luxury commodity reserved for economic winners.

Can democratic institutions scale what functions effectively at Exeter? Dramatically reduced class sizes, teachers permitted to exercise legitimate intellectual authority grounded in knowledge and judgment rather than coercive power through standardized curriculum and grades, seminar-based dialogical instruction, mentorship relationships rather than crowd-management protocols?

Or will the nation accept a rigorously bifurcated system: Harkness tables led by authentic intellectual authorities for the affluent, algorithmic compliance training administered through corrupted grading systems for everyone else?

Arendt insisted that authentic education requires adults to assume comprehensive responsibility for the world they bequeath to successive generations. “Anyone who refuses to assume joint responsibility for the world should not have children and must not be allowed to take part in educating them.”

The artificial intelligence panic lays bare this fundamental contradiction. We standardized the very practices we claimed to value most highly while forbidding teachers from asserting that anything genuinely matters beyond evaluative scores. We subjected children to peer tyranny by removing authentic adult relational authority while installing corrupted power structures disguised as objective assessment.

Artificial intelligence did not precipitate this crisis. It merely rendered our collective bad faith impossible to sustain through continued willful ignorance. The choice confronting American education could scarcely be more stark.

Either the nation commits to genuinely educating all children well, which necessarily entails both thoroughgoing structural transformation of how schools operate and fundamental reconception of teacher authority as legitimate rather than as coercive power exercised through grading, or we acknowledge that democratic education always constituted a mendacious fiction and permit voucher schemes to complete what ChatGPT initiated: the final abandonment of education as shared inheritance in favor of education as market-distributed privilege reserved for those few communities where adults still exercise legitimate authority rather than corrupted power.