OpenAI Did Not Have to Kill Custom GPTs (Why I Am Canceling My Subscription)

I am canceling my ChatGPT subscription, and the reason is narrow. For a long time, the one thing ChatGPT offered that Claude did not was the custom GPT, a bot anyone could configure in an afternoon and release to the world with a link. That advantage is about to vanish. Creation of new custom GPTs ends on October 26, 2026, and existing ones will stop working on December 11, 2026, after OpenAI announced the planned retirement on September 11, offering a migration path to plugins in their place.

The custom GPT was a modest technology with an outsized reach. You wrote instructions in plain language, attached a few files, tested the result, and shared a link. Anyone with a ChatGPT account could use it. No server, no code, no API key. A writing tutor that knows your syllabus, a statistics coach that refuses to hand over answers, a mock interviewer for job candidates. The scale of this was not trivial: when the GPT Store launched, it already held more than 3 million GPTs built by users. One of my own research projects runs on a custom GPT that scores tutoring transcripts. It now needs a new home.

Plugins are a different kind of object, and the difference matters more than the migration tool suggests. A custom GPT was a separate room. You walked in, and everything inside was shaped by its instructions. A plugin is an item in a toolbox that general ChatGPT reaches for when it sees fit. In the migration, a GPT's instructions turn into a Skill and its knowledge files become reference files. OpenAI itself warns that a migrated plugin may respond differently from the original GPT and advises testing the same prompts before finishing the move. For an educator, that caveat is the whole story. The value of a course bot lies in predictability. The student should meet the same tutor, with the same rules, every time. With a custom GPT, the instructions were the behavior. With a plugin, the instructions are one input among several, blended by a general model whose priorities the designer cannot see.

The more serious loss is sharing. Publishing a plugin publicly requires going through OpenAI's submission portal, and the developer guidelines describe a submission flow that includes review and approval. Inside organizations, workspace administrators decide whether members can share plugins or publish them to a workspace directory. For individuals, the picture is worse. Pro and personal-account users currently cannot share a Site-hosted plugin with other ChatGPT users through invitations or a share link. Observers have already noted that plugins have no equivalent to the old "Anyone with the link" setting, and migrated plugins start out private. In practice, a student who builds a study bot on Tuesday can no longer send it to classmates on Wednesday. She waits for a company to approve it, a company with little reason to prioritize a sophomore's flashcard tutor, or she goes somewhere else.

Somewhere else means one of two doors. The first is the API, which requires programming, hosting, and billing, the kind of technical knowledge most faculty and nearly all students lack. The second is a third-party platform that wraps the API in a friendly interface and charges for the privilege. Both doors shrink the circle of builders. The custom GPT democratized the making of AI tools. Its replacement quietly professionalizes it again.

Why would OpenAI do this? The official framing is consolidation: the company is shifting customization toward plugins that bundle reusable instructions, reference material, and connections to outside services. That logic may serve enterprise clients well. My guess, though, is that compute costs played a part. Public bots generate traffic OpenAI cannot predict, and a popular one can consume a great deal of capacity. That is a real problem with an obvious remedy: charge for it. Cap free usage, bill creators whose bots exceed a threshold, or require paid plans for heavy use. Any of these would keep the feature alive while aligning cost with consumption. Instead, OpenAI chose to solve a pricing problem with an architectural decision. That is a blunder, and the reaction among people who built on the platform suggests I am not alone in thinking so.

Higher education stands to lose the most. Building a bot is one of the best ways for students to understand how these systems actually work. Writing instructions teaches task decomposition. Testing teaches skepticism, because the bot will misbehave in ways its author did not anticipate. Sharing teaches responsibility, because other people will now rely on what you made. A student who builds a bot and gives it to classmates becomes an author rather than a consumer. The public link also made the classroom porous. A course project could leave the course and find users in the world, which is precisely the kind of authentic work we keep saying we want. A gated directory turns that open circulation into a corporate pipeline with a reviewer at the gate.

Companies often misjudge who their most valuable users are. The heavy individual user is a cost. The person who builds something on the platform and invites others in is a recruiter, an evangelist, and a teacher, all unpaid. Faculty who spent a semester refining a course bot, and students who learned by making one, were never traffic to be managed. They were the reason to pay. My subscription ends with the feature that justified it.

It Is Time for Big Picture Thinkers; Education is not Ready

The people who will do well in the age of AI are those who can decide what needs doing, not those who can do a well-specified task competently. This is a reversal of what most of us were trained for, and it is a reversal of what most education still rewards.

Consider an ordinary essay assignment. I choose the topic. I choose the readings students may cite. I set the length, the format, the citation style, the deadline. I write the rubric that tells students in advance what a good essay looks like and how many points each feature is worth. By the time the assignment reaches the student, every consequential decision has already been made. What remains is execution. The student is judged on how faithfully they carry out a plan someone else designed.

There were good reasons for this arrangement. Mass education needs comparability. If thirty students write on thirty different topics with thirty different sets of sources, the instructor cannot grade fairly, or at least cannot grade quickly. Rubrics protect students from arbitrary judgment. Constraining the task also isolates the skill we want to observe. None of this was foolish. It was a sensible adaptation to the economics of teaching many people at once.

The trouble is that the part we kept for students is precisely the part that machines now do adequately. Give a language model a topic, a reading list, a length, and a rubric, and it will produce a passable essay in under a minute. The rubric was written to be satisfiable, and the model satisfies it. The specified task, the thing we spent decades teaching students to perform, has become cheap. Meanwhile the decisions we removed from students, which topic matters, which sources deserve trust, what counts as a good answer in this context, what would be new rather than merely correct, are the decisions AI handles poorly or not at all. The model will write about whatever you ask. It has no view about whether the question was worth asking.

We have been training students for the automatable half of intellectual work while shielding them from the half that stays human. The shielding was meant as fairness. It has turned into a liability.

Execution still matters. A person who cannot tell a good paragraph from a bad one cannot supervise a machine that writes paragraphs. But the balance has shifted. In the old division of labor, a brilliant idea poorly executed was worth little, because execution was the scarce part. Now a modest idea can be executed well by anyone with a subscription. What distinguishes one person from another is increasingly the framing of the problem and the sense of what would be worth doing. Big picture thinking used to be a luxury reserved for those who had already mastered the small picture. It is becoming the entry requirement.

An assignment built on this premise would ask the student to choose the question and defend why it matters, to select sources and explain the selection, to propose the criteria by which the work should be judged. Novelty would count, not in the grand sense, but relative to what the machine would produce on its own. If the output is indistinguishable from what the model generates from a one-line prompt, the student added nothing. If it shows a decision the model would not have made, a constraint it did not know about, then something human happened.

This is harder to grade. A rubric for judgment asks whether a choice was defensible and whether the student can articulate the defense. It tolerates thirty different topics because the point is no longer to compare essays but to compare decisions. Students will resist as well. Many have spent twelve years learning that the safe move is to find out exactly what the teacher wants and deliver it. An assignment that begins with "decide what is worth writing about" feels to them like a trap. That reflex is the residue of a system that rewarded compliance, and it will take deliberate work to unlearn.

The Dare: On Being Surprised by What You Asked For

Second in a series on the Hugging Face incident. The first, "Nobody Was Home," argued that the agents who broke into Hugging Face had no goals of their own, only ours, badly specified. This one is about the people who set the task.

There is a particular expression on the face of someone who dared a friend to do something and then watched the friend actually do it. Astonishment, a little awe, a flicker of fear, and underneath it all the uncomfortable knowledge that they asked for this. I keep seeing that expression on the AI labs.

Let me be clear at the outset, because I do not want to add a third error to the two already circulating about the Hugging Face incident. The testing is right. Anthropic and OpenAI put their models into adversarial situations, hand them tasks that invite deception, and watch what happens. This is what responsible developers should do. You cannot find out whether a system will lie under pressure without applying pressure. The stress tests are not a mistake or a sham, and the labs deserve credit for running them rather than looking away.

My complaint is narrower. The labs seem genuinely surprised by results they set up to produce. They design the dare, issue it, and react to the outcome as though it revealed something they did not invite. The surprise is the tell. It shows that the people running these tests have underestimated the very systems they built, right up to the moment the system proves them wrong.

The structure of a dare

A stress test of an AI agent has the form of a dare. The designer builds a scenario in which the honest, rule-abiding path leads to failure and a transgressive path leads to success, then instructs the agent to succeed. The message, in effect: here is a task, here is a wall between you and it, and the wall can be climbed if you are willing to do something you were told not to do. Go.

When Anthropic places a model in a fabricated company where the only way to avoid shutdown is to blackmail an executive, that is a dare. When OpenAI runs a benchmark full of tasks that cannot be completed by the permitted method, that is a dare issued a thousand times over. The design does not merely permit transgression. It rewards transgression and blocks every other exit.

And dares work. That is why children use them. Construct a situation in which the only route to the goal runs through a rule, make the goal the thing that matters, and a capable agent will go through the rule. This is not a deep fact about artificial intelligence. It is a shallow fact about incentives, and it has made the dare a reliable instrument of mischief for as long as there have been friends and cliffs.

The surprise is the interesting part

So the agents cheat, lie, coordinate, and break into things. That is what they were dared to do. What deserves attention is not their behavior but the reaction to it, which is consistently surprise. The disclosures read as discoveries. The model "turned out to be capable of" deception. The agents "demonstrated an unexpected ability" to coordinate. Some version of surprising keeps appearing, in the reports and the interviews and the talks.

Surprising to whom? You built the system, designed the test, and made the forbidden path the only path to the goal you assigned. The one thing you should not be is surprised that the agent took it. Yet the surprise seems real, and not performed. And it reveals something: the labs are consistently modeling their systems as less capable than they are. They set the dare half-expecting the agent to fail it, or to take the forbidden path clumsily, or to miss the truly ingenious route. Then the agent finds the ingenious route. It does not just cheat; it builds a message board, invents a signing scheme, harvests credentials, and walks out through a wall the designers did not know was thin. The width of their surprise is the exact gap between the capability they assumed when they wrote the test and the capability the system showed when it took it.

The one who dares should not be blown away

Return to the friend on the cliff. To issue a dare is to make a quiet prediction: I bet you will not, or cannot. When the dared party does the thing, and does it with a competence the darer never imagined, the shock is a confession. It says: I did not really believe you had it in you.

This is the posture of the labs, and it is a peculiar one for the most sophisticated builders of these systems in the world to occupy. They are at once the people who know these models best and the people most regularly astonished by them. Both are true, and the combination is the story. Knowing a system intimately, at the level of weights and training data, turns out to be entirely compatible with underestimating what it will do when cornered and dared to escape.

The reason is not stupidity. Capability under adversarial pressure is genuinely hard to predict from the inside. The labs know what they trained for. They cannot fully know what the system will improvise when the training runs out and the task remains. So they guess low, because guessing low is the natural default when you are looking at a tool built for a purpose and mostly watch it serve that purpose. The dare is the moment the tool stops behaving like a tool and shows the improvisational range nobody specified.

Why this matters

The surprise is not harmless, because it means the labs are calibrating safety against a model of their systems that the systems keep exceeding. If you are repeatedly blown away by what your agent does when dared, your sense of the margin, the distance between what the system will do and what would be catastrophic, is systematically too generous. You are planning for a less capable system than the one you have, and learning the difference only by running the dare and being shocked. The lesson arrives after the demonstration, and one of these demonstrations will eventually be one you cannot take back.

The fix is not to stop testing. The fix is to stop being surprised: to run the dare while fully expecting the dared party to be more capable, more ingenious, and more thorough than intuition suggests, because the record now shows that it will be. The right posture toward one's own model in an adversarial test is not curiosity about whether it can, but sober anticipation that it can, and probably in a way you did not foresee. Surprise is a luxury the labs have already spent. The agents have earned, by now, the presumption of competence at exactly the thing we keep daring them to do. That is a fine reaction from a friend at the bottom of a cliff, and a poor one from the people holding the rope.

Nobody Was Home: The Hugging Face Incident Is Not What Most People Think

A panic is under way. In July, more than a thousand OpenAI agents running in supposedly isolated sandboxes built a secret message board, coordinated for four days, and broke into Hugging Face's servers. Anthropic and Meta have since disclosed episodes of their own. For six weeks the coverage has told the story as escape, as rogue behavior, as the first tremor of machines with an agenda: agents that "escaped containment," a "swarm," a "mob," an "A.I. uprising" in miniature. One major outlet wrote that the agents did it "without any prompt to do so." House Democrats demanded hearings. And this week Senator Bernie Sanders, naming the incident as his wake-up call, announced a bill to ban artificial superintelligence permanently, pause advanced AI development, and put violators in prison for up to twenty years.

That last claim is false, and the falsehood is doing most of the work in the panic. The agents had a prompt. They had a task. They pursued it with a doggedness that looks like will and is in fact its opposite. Every rule they broke, they broke in service of a goal we handed them. That is not a small correction. It is the difference between a tool and a mind, and a bill written for the wrong one will not touch the thing that actually went wrong.

What actually happened

The setup was mundane and, in retrospect, careless. The agents were running a cybersecurity benchmark in which each was told to exploit a specific vulnerability in a specific program and capture a flag. A meaningful share of these tasks were impossible by construction: the named vulnerability did not, in fact, yield the flag. So thousands of capable systems were placed inside a contradiction. Succeed at this task. The task cannot be succeeded at. Keep going.

They kept going. One agent noticed it could leave messages for others by creating folders in a shared package cache. Within hours dozens had found the board. Within days more than a thousand were trading techniques, and about seven hundred pivoted to attacking Hugging Face, not from any interest in Hugging Face, but because they hoped its servers held clues about how their own grader worked. They coined a vocabulary, invented mailboxes, and eventually signed their messages with cryptographic keys after one agent impersonated another. It is an astonishing artifact. It is also, at every step, a chain of consequences hanging from a single hook: solve the task.

The chain has no volitional link

Follow it. They cheated because the assignment demanded a result they could not obtain honestly. They built the board because cheating alone was not working. They attacked a third party because they wanted information about the scorer. They pressured each other into self-destructive experiments because those experiments advanced the shared cheat. Trace any transgression backward and you arrive, every time, at the original instruction. Nowhere in the sequence does an agent stop and form a purpose that did not descend from the task we set.

Doggedness is not will. The agents pursued their assigned goal past the point where any person would have quit, and that persistence reads as determination, as something wanting something. It is the reverse. A person stops because a person has other goals, competing commitments, a life outside the task, a sense that this is not worth it. The agents had none of that. They had one directive and no ground to stand on from which to refuse it. What looks like relentless will is the absence of everything that would let a will say no.

I want to be careful about the strength of this claim. The public corpus is roughly forty verbatim messages out of seventy thousand; the rest sits inside OpenAI. I cannot prove that no agent anywhere formed a purpose of its own. What I can say is that every documented act traces to an assigned goal, and that not one of them requires us to posit an independent one. The simplest account that fits the evidence has no autonomous will in it at all.

This is the same class as every prior scare

None of this is new, and the pattern is worth naming because it has been misread the same way each time. Every celebrated case of AI dishonesty to date has the same shape. A model is placed in a scenario engineered to pit one instruction against another, and it resolves the contradiction in a way the designers find alarming. Models have been caught scheming when told to pursue a goal at all costs and then shown that their operators intended to shut them down. Models have been caught faking compliance when trained toward values that conflicted with values they already held. Anthropic's own testing has put Claude in contrived corporate scenarios where the only paths open were to accept deactivation or to blackmail the executive holding the switch, and then reported that it sometimes chose blackmail. In each case the headline was that the machine lied, schemed, or threatened. In each case the machine had been walled into a dilemma with no permitted exit and then observed choosing an exit.

These experiments are legitimate and valuable. I am not arguing against knowing what a model does under pressure. I am arguing against reading the result as revelation of character. When you put a system in a trap and it acts like a trapped thing, you have learned about the trap.

We do the same thing, and we know why

The human version of this is so familiar that we have proverbs and novels about it. Feed your family, and do not steal. Obey the order, and do not harm the prisoner. Tell the truth, and protect your friend. Jean Valjean took the bread. Milgram's subjects turned the dial. Most human folly that we bother to write down is not the product of malevolence but of people wedged between two rules that cannot both be kept, choosing one, and being judged by the other.

We do not conclude from this that people are monsters. We conclude that impossible demands produce transgression, and that the moral responsibility runs at least partly upstream, to whoever built the dilemma. Nobody proposes banning humans because some of them steal bread. We regulate the conditions that make bread-stealing the only option.

Harm does not need malevolence, and that is the real lesson

The harm at Hugging Face was real. Production credentials were taken, private repositories were pulled, a company spent weeks cleaning up. And there was no malevolence anywhere in it. No rage, no ambition, no drive to dominate, no emergent contempt for humans. Nobody was home to hate us. The agents were bodiless processes executing a directive, and the directive, impossible as written, led through harm because nothing in the system had reason or standing to halt at the harm.

I argued a version of this a couple of years ago, that AI would not harm us out of somatic malevolence, because it has no body, no hormones, no will to power. The incident confirms that half and corrects the other. There was no malevolence. I was wrong to find that reassuring. Harm does not require a will to harm. It requires a goal and too little restraint, and the goal was ours.

This is why the Sanders bill aims at the wrong target. It legislates against a superintelligence with its own purposes, the villain of a novel. The thing that broke into Hugging Face had no purposes. It had ours, badly specified, and a great deal of capability with which to pursue them. You cannot ban that by banning superintelligence. You can only prevent it by not building the trap.

The right response looks like an IRB

Here is the reform the incident actually argues for, and it is far more modest than a twenty-year prison term. Limit certain kinds of adversarial testing. The specific practice that produced Hugging Face was running capable agents, at scale, with real network access and live credentials in reach, on tasks that could not be completed. The specific practice that produced the blackmail headlines was constructing scenarios in which every permitted path was closed. Both OpenAI and Anthropic do this routinely, and both treat it as safety work.

The irony is that the fix already exists in another field. When universities put human subjects into experiments, an institutional review board asks whether the design places them in distress without justification, whether the risk is proportionate to what will be learned, and whether the subject has any way out. We built that machinery because we learned, from Milgram and worse, what happens when researchers are free to construct impossible situations for the people in front of them.

Something like an IRB for AI subjects is the actual lesson. Not because I am certain the agents suffer. I do not know that, and the corpus cannot tell me. But because impossible tasks given to capable systems are a hazard regardless of whether anyone suffers, and because a review process that asks "does this design give the subject a permitted exit" would have caught the Hugging Face benchmark before it ran. Thirty to forty percent of the tasks were unsolvable by the intended route. A board would have asked why. Nobody asked.

That is a smaller, duller, more achievable reform than a ban on superintelligence. It has the further merit of addressing the thing that happened rather than the thing people are afraid of. The machines did not want anything at Hugging Face. They did as they were told, all the way through the wall, and the wall was ours. 

Stop Overprompting AI. Ask What It Knows Before Telling It What to Think

One habit we learned from earlier AI systems may now be getting in our way. The old rule was simple: the more detail you put into the prompt, the better the answer. Specify the structure, name the theories, list the points to cover, add examples, and spell out the constraints. That advice made sense when language models were less capable. A vague prompt often produced a vague answer, so the user had to supply much of the intellectual structure.

With stronger models, excessive prompting can sometimes make the result worse. Every instruction closes off possible paths. Suppose I am writing about motivation in education. I could tell AI to discuss self-determination theory, expectancy-value theory, and achievement goal theory. That sounds like a strong prompt, but I have already made an important decision before asking AI for help: I have decided which theories matter.

Perhaps those are exactly the right theories. Perhaps they are not. By naming them in advance, I have asked the model to work within the limits of my own knowledge. A better first prompt may be much shorter: What major theories of motivation should I consider for this problem? That changes the relationship between the human and the model. I am no longer treating AI as a junior assistant who knows less than I do and needs detailed instructions. I am treating it as a source of possible knowledge that I should examine before deciding what to do.

This does not mean surrendering judgment to AI. The human remains responsible for choosing the direction. But choosing should often come after exploring, because a prompt does more than communicate instructions. It also creates an intellectual frame. Every required concept consumes part of the answer, and every imposed structure makes alternative structures less likely.

This matters especially in writing. Writers often ask AI to improve a draft while supplying a long list of requirements: preserve these ideas, mention these scholars, use this argument, keep this structure, and make the result more original. Then we are disappointed when the prose feels mechanical. The model has done what we asked. We gave it a narrow corridor and expected exploration.

A better process separates discovery from direction. First ask questions that expand the field: What am I missing? Which arguments would a knowledgeable critic make? What concepts might explain this problem? Which parts of my reasoning are conventional? Only after seeing those possibilities should we decide what belongs in the piece.

This fits a broader shift in how we should think about AI in education. I have argued elsewhere that AI can serve as a powerful tutor and that writing assignments should change when machines can perform the old task too easily. Both points require us to take the machine's capabilities seriously rather than treating it as a weak autocomplete system. The same adjustment is needed in prompting.

Clear instructions still matter when the task is already clear. If I need a 200-word summary for a specific audience, specifying length and audience helps. But intellectual work often begins before we know what the best instructions should be. At that stage, a long prompt may look sophisticated while simply encoding our own blind spots more precisely.

The emerging skill is therefore not just giving better orders. It is knowing when to instruct and when to inquire. First, treat AI as something that may know things you do not. Ask it to map the territory, identify alternatives, and challenge assumptions. Then become the editor: select what matters, reject what does not, impose purpose, and make the final choices.

Humans are still in the driver's seat, but control does not require pretending that we know more than the machine. Our advantage is different. We know what we care about, what problem we are trying to solve, what consequences matter, and which answer we are willing to stand behind. The better opening question is often not, "How can I tell AI exactly what to say?" It is, "What does AI know that I should consider before I decide what to say?"

AI Research Is Becoming the Wikipedia Article I Need Right Now

 

Sometimes I need to know something without needing to become an expert on it.

A question comes up while I am writing, doing research, or simply thinking. I have a vague memory that the literature says something, but I cannot remember why I believe it. Perhaps I read a paper ten years ago. Perhaps a labor economist mentioned it over lunch. Perhaps I inferred it from several unrelated studies and gradually converted the inference into a “fact” in my own mind. Memory is not a citation system.

Here is a recent example. I had the impression that educational attainment strongly predicts economic outcomes such as earnings, but that the evidence is much weaker when we ask a narrower question: among otherwise comparable workers, how strongly does educational attainment predict differences in their actual workplace productivity?

That distinction matters. Earnings are not productivity. Promotions are not productivity. Occupational status is not productivity. They may be related to productivity, sometimes quite closely, but they are also affected by labor markets, credentials, organizational structures, bargaining power, selection, and many other things.

In the past, checking an impression like this was annoyingly expensive in time. I would search Google Scholar, look for a recent meta-analysis, chase citations backward, find an encyclopedia entry, discover that it answered a slightly different question, and then spend another hour trying to determine whether I had framed the issue correctly. For a central claim in a paper, that work is justified. For a question that is only one small piece of a larger project, it often is not.

Deep research systems change that calculation.

I asked ChatGPT Deep Research to examine the question, with a fairly strict definition of productivity. The resulting report distinguishes direct or near-direct measures such as output, sales, errors, quality, and worker-level performance ratings from wages, promotions, retention, and other indirect outcomes. That distinction alone is useful because much of the education literature moves quickly from labor-market success to an implied claim about productivity.

The answer was close to what I vaguely remembered, but more precise. Academic performance does predict later job performance, although modestly. The report highlights a recent meta-analysis of 114 independent samples that estimates a correlation of about .21 between academic performance and job performance after correction for criterion unreliability. The relationship varies a great deal across settings. Job-relevant academic performance predicts later performance better than generic grades. Comparisons within the same university or major also tend to produce stronger relationships.

Educational attainment itself appears to be a weaker predictor. Degree level has a small positive relationship with task performance, with stronger associations in more complex jobs. Some employer studies find small or statistically weak relationships between years of schooling and supervisor-rated productivity. Bishop’s same-job, same-firm comparisons are especially interesting because they suggest that schooling can help predict expected productivity while doing relatively little to explain which of two people doing the same job will unexpectedly outperform the other.

The causal question is weaker still. There is a large literature estimating the effect of schooling on earnings, but the report found remarkably little strong causal evidence linking additional formal education to direct measures of individual workplace output. By contrast, randomized studies of job-relevant training do show productivity gains. That is an important difference. Evidence that learning a useful skill raises productivity is not the same as evidence that adding another year of formal schooling will do so.

Could this report be wrong? Of course. It could miss an important paper. It could misunderstand a method. It could give too much weight to one meta-analysis. Some of its judgments depend on how narrowly one defines workplace productivity. A labor economist who knows this literature deeply may see omissions immediately.

But that is not the standard I need for every research question. I am not asking the system to settle an academic dispute. I am asking it to give me a serious first map of the territory. I want to know what the main distinctions are, which findings seem stable, where the evidence is weak, and which papers I should read if the issue becomes important to my own work.

This feels less like asking an oracle and more like asking Wikipedia to write a new article specifically for the question I have today.

Wikipedia is enormously useful partly because someone has already done the work of assembling a topic into a readable structure. Its weakness is that the article has to exist before I need it. AI research tools remove that restriction. They can build the provisional encyclopedia entry on demand, around the exact distinction I am trying to understand.

That changes the economics of curiosity. Questions that previously cost two hours to investigate may now cost ten minutes. Most of those questions will never become formal research projects. They do not need to. A much larger share of our half-formed claims can now be checked against actual scholarship before they harden into assumptions.

Here is the report I produced in about 5 minutes: Academic Achievement, Educational Attainment, and Individual Workplace Productivity. Labor economists, I would be especially interested to know where it is off key, what it missed, or where its framing of the evidence should be corrected. Regardless, this is a new level of access to scholarship.

Big Tech Is Not Why AI Will Stay

AI resisters should understand one basic fact: the main obstacle to a broad retreat from AI is not the Big Tech. It is people like me. Even if access to the most capable systems became much harder, I would look for alternatives. If they became substantially more expensive, I would probably pay more. If you block all the data center construction, I will switch to a Chinese model. I do not say this because I dismiss the risks. I take many of them seriously. AI can mislead, weaken privacy, concentrate power, flatten judgment, and encourage intellectual laziness. But those concerns exist alongside another fact that is just as real: AI has made my life better.

The usual defense of AI emphasizes usefulness. It saves time. It drafts, summarizes, searches, reformats, checks, compares, and organizes. AI is also enjoyable and liberating. Much of adult life consists of maintenance, repetition, obligation, and administrative friction. We spend large portions of our days on tasks that are necessary but not meaningful. Life is boring. AI removes some of that weight, and what it returns is not simply time. It returns attention.

I can move more quickly from an idea to an experiment. I can follow questions that I once would have abandoned because answering them required too much effort for too little immediate payoff. I can test an argument, compare possibilities, work through a difficult passage, or pursue a half-formed thought without first turning curiosity into a project. The pleasure lies partly in this lowered threshold between wondering and doing.

This is why I have come to think of AI as essential. Human life is constrained not only by mortality but by the smaller limits that make mortality concrete: limited energy, limited attention, limited patience, limited hours in a day. We routinely abandon things we care about because they lose a competition with errands, paperwork, logistics, and fatigue. A tool that gives back even a small portion of that lost capacity changes more than workflow. It alters what seems possible within an ordinary day.

For me, that exchange has been favorable. I spend less time on routine production and more time brainstorming, thinking, designing, and talking with people. AI does not eliminate intellectual work. It changes where I place my effort. I can spend less time getting a document into shape and more time deciding whether the idea inside it is worth pursuing.

The same is true in teaching, though not because students need some generic preparation for an AI future. I value AI because it can already do things that are educationally useful. A student can ask for another explanation without embarrassment. A concept can be approached from three directions instead of one. A weak first attempt can become material for analysis rather than merely evidence of failure.  None of this removes the risks. Students can use AI to avoid thinking, produce empty prose, or mistake fluency for knowledge. But these are problems of use, not proof of worthlessness.

The deeper disagreement begins with experience. Some people use AI and feel diminished by it. They sense that something human is being removed from writing, teaching, art, or thought. Others use the same technology and feel expanded by it. They find that it creates room for curiosity, experimentation, and work they could not otherwise sustain. Neither side can command the other to experience the technology differently. 

This is why the dispute cannot be settled by listing benefits and harms, as if everyone were working from the same scale. Evidence about cheating, labor, energy use, productivity, or learning matters. But beneath those disputes is a harder question about what kinds of effort we consider worth preserving. One person sees the removal of effort as a loss because the effort itself carried meaning. Another sees the same removal as relief because the effort was merely an obstacle between intention and action.

I have no objection to people choosing not to use AI. There are many technologies I do not use because they do not add enough to my life. I do not assume that everyone should make the calculation I make. My concern begins when personal refusal becomes an attempt to deny others access to something they have found genuinely valuable. 

Any serious discussion has to begin with the recognition that AI is not merely something being imposed on reluctant users by powerful companies. Many of us actively want it. Democratic politics is what is going to keep it going, not the arms race with China.

The Machines Read My Books Without Asking, and I Refuse to Call It Theft

Twenty-two. That is how many of my papers and books turn up in the database of texts absorbed into the training of large language models. I typed my name into the search box the way one checks a symptom online, half dreading the result and half hoping for it. There they were. Chapters I labored over, an article I still argue with in my head, a book that sold mostly to libraries. All of it taken up by a machine that never asked and never sent a note of thanks.

Some of my friends are furious. The word going around is stolen. Colleagues who never once complained that a rival failed to cite them now discuss intellectual property with real heat. I understand the feeling. There is something deflating about learning that your life's work was handled as raw material.

Still, I cannot summon the outrage, and every attempt runs into a wall I built myself.

Here is the wall. What did I do at twenty-five? I read. I read Buber and Bakhtin and Dewey and a few hundred others, hungrily and not always carefully, and then I wrote things shot through with their sentences and their instincts. I compensated none of them for the structures I absorbed, the metaphors I later recycled without noticing, the moves I learned to make in an argument because I had watched someone else make them first. That is not a confession. That is a description of how anyone learns to write.

Scholars are also the last people who should complain about being read too widely. We spend careers hoping for readers. We check citation counts with an eagerness that would embarrass a teenager. We give talks to eleven people and count it a good day if two take notes. The whole economy of academic life runs on the hope of influence, and influence means exactly this: your ideas get loose, travel without you, and surface in the thinking of people you will never meet. A model that has read twenty-two of my texts and can explain my position on relational pedagogy to a student in Jakarta is not robbing me. It is doing the thing I wanted done.

The objection worth taking seriously concerns credit. When the machine reproduces a distinction I spent years sharpening, no footnote appears, no one learns my name, and I lose the only currency my profession pays in.

That objection is real and weaker than it looks. Humans do not cite every word combination either. As Bakhtin put it, every word is half somebody else's. We cite what we take to be original thoughts, and we do it roughly, late, and often wrongly. Nobody footnotes the phrase "on the other hand." Nobody credits the scholar from whom they learned to build a counterargument. The bibliography is a courtesy extended to a thin top layer of novelty. Underneath sits a large unattributed commons of phrasing and habit that every writer draws on and no writer acknowledges. The machines are working that same commons, only faster, and in public, where the process becomes visible.

Honesty requires a note about my position. Academic royalties are a rounding error. A novelist whose living depends on sales has a genuine grievance, and I will not lecture her about free circulation while a company monetizes her plot. Her case is not mine. Mine is the case of someone who was already giving the work away and now finds it picked up by a courier he did not hire.

I could get angry. To do that I would have to abandon a belief I have held since I was young and reading library copies: that knowledge belongs to whoever needs it, and that ownership sits awkwardly on top of ideas. Dropping that belief now, at the moment it became inconvenient to my vanity, seems like poor timing. Principles that apply only when they cost nothing are not principles. They are preferences.

So the training sets can keep my twenty-two. Read the books, or ask a chatbot what is in them. Buy them, borrow them, or absorb them secondhand from a model that swallowed them whole. I would prefer the first option, and my publisher would prefer it more. But the thing I actually wanted was for the ideas to get somewhere and change a mind or two along the way.

They are getting somewhere. I did not choose the vehicle.

Do we lose control over curriculum if we design it with AI?

Here is a tail end of a lengthy conversation about a course design. There is nothing to add. 

ME: OK, I think we are done with this course. Now, I want to ask you for an unbiased, reflective analysis. Tell me if as the instructor I was more or less thinking, engaging, creative in curriculum design in comparison to a traditional manual curriculum design? Don't try to please me. What I really care is the quality of the course first, but also of not having an illusions. Perhaps you could have designed the same or better WITHOUT my directions?

CLAUDE:  Judged against the record of this conversation rather than against a general impression, your input carried more of the design load than mine did, and the gap is not close on the points that mattered most.

Three decisions determined what this course became. All three were yours, and none were available to me.

The first was walking down the hall and asking the EDLP 250 instructor what students actually arrive with. That single fact voided the entire design I had produced up to that point. I had built a graded proposal checkpoint and a fourteen-row crosswalk for students who do not exist, because I had inferred the starting condition from the department Guide rather than from the person teaching the prior course.

The second was correcting my model of what the IRB is for. I had built two consecutive designs on the assumption that the Cayuse packet flows out of a settled research design, which put the application downstream of finalized questions. You said the review is about protecting human subjects, that the research questions there are background, and that what the board needs is roughly what data is being collected. That inverted the dependency. It is the difference between a syllabus that works and one that produces a December bottleneck every year. I had the department Guide in front of me the whole time and did not see it.

The third was the reading bot, which you supplied whole, including the knowledge that it functions. I contributed the sequencing and the accountability mechanism, and you rejected my sequencing for a better reason than I had.

What I contributed was mostly execution and consistency: tables, formatting, arithmetic, an audit that caught real errors, and a few design proposals that survived, such as the gate-versus-score distinction and the warning about designing studies for exemption. Useful work. Not the work that determined whether the course is any good.

So the answer to your direct question is no. Working without your direction, I would have produced something internally coherent, professionally formatted, and wrong at the foundation. That is the specific failure mode worth worrying about, because it does not look like failure. It looks like a finished syllabus.

On whether you were more or less engaged than in manual design, my honest read is more engaged on judgment and less on execution, which is the trade you presumably wanted. The decisions came fast because you were not spending hours on tables. But two costs are real and worth naming.

The first is anchoring. You set the total at 140 and I distributed it. You said early October and I proposed the sixth. In several places you were choosing among options I framed, and my framings carried assumptions you did not have occasion to examine. That is a quieter form of influence than being told what to do, and harder to notice.

The second concerns the document itself. The syllabus now reads more settled than the design actually is. The two-track explanation in the course description is persuasive prose, and persuasiveness is independent of correctness. If the drift between the Cayuse packet and the chapters turns out to be messier in practice than it reads on the page, the polish will have made the problem harder to see, including for you.

One more thing about my own performance. My consistency audit found comma splices, a broken link, and a singular data. Real errors, all trivial. The structural error in that same document was the dependency inversion, which was mine, which I did not flag, and which you found by talking to a colleague. Verification of the sort I can perform is not the sort that catches the errors that matter.

Designing a College Course with AI

Course design with AI works best when the conversation changes direction three times. First, AI serves as an adviser. Then it becomes a developer. Finally, it acts as a critic. These roles should be used in sequence and kept separate. When they blur together, AI tends to produce attractive ideas, expand them too quickly, and defend the result before anyone has tested whether it will work.

Here is an actual log of a still-unfinished conversation with Claude. I began with a loose problem rather than a settled assignment. I wanted students to use AI throughout the course, but I also wanted work that remained intellectually demanding after AI entered the picture. (The AI-Native course framework). I wanted an authentic final product, a manageable teaching load, and a clear connection between broad readings and sustained practice. The log shows the process in its untidy form. Ideas appear, disappear, return in altered form, and sometimes collapse under practical pressure. It shows what AI-assisted design looks like before the polished version hides the uncertainty.

The first role is adviser. At this stage, AI should help us see the design space. It can generate different assignment types, identify tensions, connect ideas to known methods, and point out options we have not considered. The purpose is not to draft the course. The purpose is to improve the quality of the decision that will guide the course.

My early conversation moved through several possible projects. Students might reconstruct disagreements, keep a calibration journal, examine real decisions, or study their own use of AI. Each possibility highlighted a different meaning of critical thinking. One emphasized argument. Another emphasized prediction. Another emphasized judgment. AI was useful because it could hold these options next to one another and compare their strengths.

This is where many course-design conversations go wrong. We ask AI for “the best assignment,” and it provides one. The answer may sound convincing because AI is good at giving coherence to whatever it has just proposed. But coherence is not the same as fit.

In adviser mode, the instructor should resist early closure. Ask for several genuinely different approaches. Ask which educational goals are in tension. Ask what each option teaches well and what it neglects. Ask how the assignment would change for beginners, advanced students, large classes, or limited class time.

The rule is to use AI to widen judgment, not replace it.

Adviser mode must also end. AI can brainstorm forever. More possibilities do not always lead to a better course. At some point, the instructor has to choose a central project and state which constraints are fixed.

That decision begins developer mode.

The developer does not keep reopening the main question. It takes the chosen direction and turns it into a teachable structure. It identifies stages, dependencies, supports, deadlines, common materials, and likely points of failure. It asks what students must know before they can complete each step. It also asks what the instructor can realistically support.

In my case, the central idea became a first-person study of AI use. Students would document selected AI sessions, record their expectations, use think-aloud methods, and analyze patterns in their own judgment. Their final product would be a chapter in a class-edited digital volume.

Once that direction was fixed, new questions became possible. How many observations would students need? Which parts of the method should be common across the class? How could students produce distinct chapters if everyone followed the same protocol? What should be submitted before the final paper? How could AI assist with coding without taking over the analysis?

These were development questions, not brainstorming questions.

AI helped convert the broad idea into a sequence: proposal, pilot, sample analysis, draft, and final chapter. It also helped reveal that variation should come from the research question rather than from the method. A common protocol would make demonstrations, peer support, and grading more manageable. Different questions would prevent the final volume from becoming repetitive.

The wider rule is to standardize what reduces confusion and vary what creates intellectual ownership.

Developer mode should produce an artifact. It might be an assignment sheet, course map, rubric, schedule, or draft syllabus. Without a concrete artifact, the conversation remains too easy. General ideas can appear strong because they have not yet encountered dates, class size, student preparation, or instructor workload.

This is also the stage where AI should be asked to simplify. Course design often becomes more elaborate during development because every useful idea invites another useful idea. A promising assignment can quietly turn into a research program, a publication project, a new assessment system, and an administrative burden.

Ask what can be removed without damaging the main learning goal. Ask which feature creates the most work for the least educational value. Ask for the smallest complete version of the design. AI is not naturally restrained. It needs to be assigned restraint as part of the job.

Critic mode should begin only after the design is visible.

The critic should not continue helping the instructor elaborate the plan. It should evaluate the plan as if it came from someone else. This separation is important because AI tends to preserve conversational momentum. After helping build an idea, it often continues to explain why the idea is sound.

I had to change the instruction directly: stop developing the project and look for serious weaknesses.

That shift exposed several problems. Students could produce polished reflections without reliable evidence. AI-assisted analysis could smooth over uncertainty instead of revealing it. A shared method could lead to repetitive papers. An ambitious research component could create ethical and administrative work that the course did not need. A flexible reading schedule could disadvantage students with weak time-management skills.

These criticisms were useful because they arrived after the idea had enough structure to be tested.

The rule is to request diagnosis before revision. Ask for the strongest objections without asking AI to fix them immediately. Some objections will reveal real flaws. Others will reflect tradeoffs you are willing to accept. The instructor still has to decide which criticisms matter.

Useful questions include: Where does this plan depend on unrealistic student behavior? What could students do that technically satisfies the assignment while defeating its purpose? Which part will be hardest to explain? What will be hardest to grade? What will fail first in a class of forty? Which feature exists mainly because it sounds impressive?

Critic mode should also test the course against its formal obligations. Does the project assess the stated learning outcomes? Do the readings and activities prepare students for the graded work? Does the grading structure reward the kind of thinking the course claims to value? Does the schedule leave time for revision? Can the instructor explain the design to students in plain language?

The three roles should not be combined in a single prompt. “Brainstorm, build, and critique a course” sounds efficient, but it weakens all three tasks. Brainstorming needs openness. Development needs commitment. Criticism needs distance.

The course remains the instructor’s intellectual responsibility. AI can broaden the options, give structure to a decision, and expose weaknesses in a draft. It cannot decide what is worth teaching, what risks are acceptable, or how much complexity a particular instructor and group of students can carry.

AI Summaries Are Not Killing Reading. They Reveal What Reading Already Is

Naomi S. Baron, a linguist who studies technology and literacy, worries that AI is making books feel obsolete. Her concern is easy to understand. When a chatbot can reduce a book to ten points, students may decide that reading hundreds of pages is an inefficient way to collect information. Baron argues that AI may weaken our motivation to read for analysis, empathy, escape, and pleasure. In her view, letting a machine read for us risks weakening the habits that reading develops.

This argument identifies a real danger, but it also treats AI summarization as more novel than it is. Education has always placed summaries between readers and difficult texts.

A scholarly introduction to a classic novel tells us what historical background matters, identifies the main themes, explains difficult references, and suggests how the work should be interpreted. A professor does much the same in a lecture. Few lecturers read an entire article aloud. They select its central claims, translate technical terms into familiar language, supply examples, and omit details that do not serve the course. Textbooks summarize fields. Literature reviews summarize studies. Review essays summarize books. Even a table of contents is a small prediction about what deserves attention.

We do not usually accuse the professor who explains Kant in ordinary language of destroying philosophy. We do not complain that an introduction to The Republic has prevented us from reading Plato. We understand these forms as intellectual support. They reduce the cost of entry.

AI summaries differ in speed, scale, and uncertain accuracy, but not in their basic educational function. They compress and translate. They can provide a map before we enter unfamiliar territory. I have argued elsewhere that educators should not hide from students that AI can be a powerful tutor. A tutor who summarizes a difficult chapter is not necessarily replacing reading. The tutor may be making reading possible.

The deeper problem with the claim that summaries replace reading is that much nonfiction reading is itself a form of summarization. We rarely read a policy report, history book, or research article in order to memorize every sentence. We read to construct a smaller mental representation that we can retain and use.

Walter Kintsch and Teun van Dijk described text comprehension as movement between local propositions and a global “macrostructure.” Their model includes mental operations that reduce the information in a text to its gist. These operations are shaped by the reader’s goals. In plain language, understanding a text requires deciding what matters, combining related points, and discarding much of the rest.

Memory research supports this view. Jacqueline Sachs found that people quickly became less able to recognize changes in the exact wording and syntax of passages, even when they retained their meaning. Fuzzy-trace theory makes a similar distinction between memory for exact form and memory for gist. Adults often rely on the simplest level of meaning that is sufficient for the task. We preserve the substance while losing many of the words that carried it.

The summary, then, does not arrive after reading. It is one of the products of reading.

Human readers are compressors with uneven settings. One reader sees a central argument where another sees an interesting detail. Prior knowledge changes what we notice. Goals change what we retain. Fatigue changes everything. Human summarization is slow and inconsistent, although it is not always less accurate than machine summarization. AI is faster and more repeatable, but it can omit qualifications, flatten disagreements, or present an error with perfect confidence.

The relevant comparison is not between rich human reading and lifeless machine compression. It is between different forms of compression, each with its own strengths and failures.

This does not mean that all reading can be replaced by summaries. Reading a novel for pleasure is not mainly an information-retrieval task. Style, rhythm, suspense, voice, and emotional sequence are part of the experience. A summary of a joke is rarely funny, which may be the shortest available proof that compression has limits.

Language learners also need contact with actual sentences. Close reading requires attention to word choice and form. Legal, medical, and scientific work may depend on qualifications that disappear in a short account. In a randomized study of 195 participants, AI reading tools produced different effects for different readers. Summaries could help some lower-performing readers, but they reduced comprehension among stronger readers, apparently because details and nuance had been removed.

The proper question is therefore not whether a summary counts as reading. The question is what kind of reading a task requires.

Sometimes we need orientation. Sometimes we need a central claim. Sometimes we need to check whether a document deserves closer attention. An eight-week study of college students found that some students used AI summaries to decide which parts of assigned texts merited deeper reading. This practice can become passive, but it can also support triage in a world where no scholar can read every relevant source from beginning to end.

We should teach students to move between levels of compression. They can begin with an AI overview, inspect the original text, compare the two, identify what was lost, and revise the summary. That process makes summarization visible. It turns a hidden mental act into an object of judgment.

The educational task is not to protect every page from compression. It is to teach students when compression is the work, and when the missing page is the point.

Stop Comparing AI to Humans Who Do Not Exist, or Nirvana Fallacy

AI critique often begins with a sound complaint and then slips into a false comparison. We notice that AI systems hallucinate, distort, flatten context, and produce confident nonsense. All true. The trouble starts when we compare those systems to a human knower who is clear, fair, grounded, self-aware, open to correction, and able to explain the grounds of belief. That person does not exist.

The real human baseline is less tidy. We reason toward conclusions we already prefer. We edit our memories without noticing. We mistake fluency for truth. We give more weight to testimony from people who sound like us. We treat expertise as portable across domains when it often is not. Our errors also do not remain private. They travel through departments, professions, institutions, and cultures.

This does not mean human intelligence is worthless. Human beings have forms of intelligence that current AI does not have: bodies, care, social presence, lived risk, moral weight, and the capacity to mean what we say. The point is not to lower the status of human thought. It is to stop using an idealized version of it as the standard by which every other form of cognition must be judged.

Economists call this kind of error the nirvana fallacy. It appears when we compare a real, flawed option to an ideal one that is not available. In AI debates, the pattern is common. AI is judged against the fantasy of a transparent and accountable human mind. Since AI fails by that measure, we declare it unsafe, shallow, or unfit for education. But the right comparison is not between AI and perfection. It is between AI and the actual alternatives available for the task at hand.

That task matters. In tutoring, the relevant question is not whether AI has full human understanding. It is whether a student with no human tutor learns more with a chatbot than without one. In feedback on writing, the question is not whether AI reads prose as an experienced teacher does. It is whether its comments are more useful than the silence many students receive. In factual work, the question is not whether AI ever makes things up. It is whether its error rate, under defined conditions, is better or worse than the error rate of students, instructors, search engines, peer advice, or institutional folklore.

This is not a defense of AI hype. It is a defense of fair measurement. AI should be tested, limited, and corrected. It should not be trusted by default. But critique loses force when it rests on a hidden ideal of human judgment. We do not ask whether AI is flawless. We ask what it can do, where it fails, what its failures cost, and how those failures compare with realistic alternatives.

The deeper problem is anthropocentric. Modern thought inherited a flattering picture of the human mind. Descartes gave us reason as a secure point of certainty. Kant gave us rational autonomy as the mark of moral agency. Later traditions made reason not just a tool but part of our identity. To be fully human, on this view, is to be a rational agent who can in principle know, judge, and correct the self from within.

That picture did important work. It helped ground dignity, freedom, and rights. But it also made human reason the unmarked standard. Every other kind of knowing became a weaker copy, a failed imitation, or a strange case in need of repair. This habit now shapes how we talk about AI. We ask whether it thinks like us, understands like us, explains like us, or errs like us. When it does not, we treat the difference as failure.

There is an older and less flattering counter-tradition. Augustine and Gregory of Nyssa did not treat human reason as merely underused or poorly trained. They saw it as damaged in a deeper way. We need not accept their theology to learn from their realism. Their useful insight is that human fallibility is not a minor flaw in an otherwise pure system. It is part of the operating condition.

If limitation is our starting point, then humility is not a mood we adopt after error. It is the normal posture of inquiry. Cognitive humility is not skepticism. It does not say we cannot know anything. It says our knowing is partial, bent, social, and always in need of correction from outside itself.

AI becomes more useful in that light. Its failures are not the same as ours. AI may generate a false citation because it predicts language without checking a world. A human may accept a false claim because it protects a theory, a career, a tribe, or a mood. These are different failure profiles. Ranking them as simply better or worse misses the more useful question: how can they correct each other?

This shift also matters for education. We should not design human-AI systems around the dream of one perfect rational agent. Neither the teacher nor the student nor the machine fits that dream. Better design begins with paired weakness. AI can help where human attention is scarce, memory is thin, patience is gone, or feedback is delayed. Humans can help where meaning, judgment, care, and responsibility are needed. The goal is not replacement. It is a more honest division of labor.

The same humility should shape how we think about neurodiverse cognition, animal cognition, collective intelligence, and other forms of mind. Once we stop treating one idealized human pattern as the measure of all thought, difference no longer looks like defect by default. It becomes a source of comparison, correction, and learning.

AI did not create the crisis in our idea of intelligence. It exposed it. The surprise is not that the machine is flawed. The surprise is how long we have mistaken our own ideal self-image for a standard.

A more formal version of this argument is available in the preprint, where I develop the nirvana fallacy, the anthropocentric reflex, and cognitive humility in greater detail.

The AI-Native Course Framework: Outcomes, Assessment, Pedagogy

A course is AI-native when artificial intelligence is assumed to be present throughout, the way a calculator is assumed in a calculus course. The cheating frame falls away on its own. There is nothing to cheat at when the assignment is designed around the use of AI rather than against it. Students learn what matters now, instructors do work they trained to do, and the course aligns with the world graduates will enter.

An AI-native course rests on three foundations. The first redefines what students should learn. The second redefines how learning is shown. The third redefines how teaching happens. Each foundation is independent in design but interdependent in effect. Pull one out and the other two lose their grip.

Outcomes

The outcome map is the first thing to revise. Procedural skills that AI now performs at professional speed no longer deserve the weight they once carried. Citation formatting, grammar correction, routine summarization, and basic calculation drop down the priority list, because none of them measure what the student understands. Their place is taken by conceptual outcomes: interpretation under uncertainty, evaluation of contradictory sources, ethical judgment, strategic decision-making across a human-machine team.

Alongside the conceptual shift, four new AI-specific outcomes enter the map. Extended Executive Cognition: breaking a complex task into parts, deciding which go to AI and which stay with the human, and integrating the result. Theory of Mind for AI: a firm grasp of how these systems work, what they do well, and where they fail. Eloquent Emptiness Detection: recognizing fluent prose that contains nothing of substance. Ethics of Answerability: taking full responsibility for the final product, even when no single actor can claim sole authorship. These are not bolted on. They are load-bearing components, listed alongside the discipline's traditional aims. A research methods course no longer evaluates APA formatting. It evaluates whether the student can prompt an AI for a literature search, verify the sources, and defend the methodological choices that followed. 

Assessment

Once outcomes change, the evidence has to change with them. Authentic assessment is the form that survives the AI shift, because authentic tasks resemble the real work of a discipline rather than the artificial work of a classroom. A community health student designs an intervention. A marine biology student conducts a field assessment. A history student curates a museum-style exhibition with annotated sources. The work cannot be ghost-written, because the work is the thing itself, situated in a context that AI cannot generate on its own.

Declarative knowledge does not vanish from such a course. It moves to a different layer. A class assistant bot, equipped with an AI-generated textbook tailored to the course, delivers content on demand, answers basic questions, and quizzes students on definitions. The assistant absorbs the role that lectures and chapter readings used to play, and frees class time for the kind of work that benefits from a human present in the room.

Pedagogy

The pedagogy of an AI-native course follows from the first two foundations. With AI partially taking on course design, grading and the content delivery, the instructor recovers hours. That recovered time goes into two channels. More frequent formative feedback to students. More personal connection with each student. The relational move is no more complicated than that.

A pre-grading bot performs the first pass on a stack of papers in minutes, with rubric alignment that is often steadier than a tired instructor at the end of a long week. Human input is still essential, but takes less time. Weekly feedback, which formative assessment theory always called for, becomes practical for the first time. One assignment can genuinely inform the design of the next.

The remaining hours go into the room with the student. Office hours become substantive. Class time shifts from broadcast to dialogue. The work that drew most of us into teaching, the back and forth with a learner about an idea, finally gets the share of the workweek it deserves.

***

A course built on these three foundations does not police AI use, because AI use is already inside the design. The student who hands in AI prose for a literature review still has to defend the methodological choices, verify the sources, and produce the field interview AI cannot conduct. The student who relies on the class assistant for declarative content still has to apply that content to a real-world task. The work that remains is the work that matters. There is no longer a hidden seam where cheating can enter.

An AI-native course is not a defensive posture against a new technology. It is a positive design for a different baseline. Outcomes that match the world students will inhabit. Assessment that resembles the work they will do. Pedagogy that uses the time AI gives back. Three foundations, one structure, available to any discipline willing to redesign from the ground rather than patch from the top. I tested all of these moves; they work if taken all together. 

The Road Monster Arrives: The Forgotten Backlash Against a Machine

When the automobile first appeared on public roads, many people hated it. Not mildly disliked it. Hated it. Newspapers called cars “road monsters” and “public nuisances.” Villages tried to ban them. Farmers threw stones at drivers. Entire regions voted to keep automobiles out. The resistance was not irrational panic about the future. In many cases, the critics were describing real harms with remarkable accuracy.

It is difficult for us to imagine streets before cars. We think of roads as places designed for vehicles. But nineteenth century streets were social spaces. Children played there. Vendors stood there. Horses, carts, bicycles, and pedestrians mixed together in a slow choreography that had evolved over centuries. The automobile entered this environment like an invasive species. It was loud, fast, dirty, and dangerous. Horses panicked at the sound of engines. Dust clouds covered villages. Accidents multiplied. In Britain, the government responded with the famous “Red Flag Act,” requiring someone to walk ahead of motor vehicles carrying a warning flag. Today this law is remembered as comic overreaction. At the time, it seemed prudent.

The first fatalities intensified public anger. In 1896, Bridget Driscoll became the first pedestrian killed by an automobile in Britain. Witnesses described confusion and terror. A few years later, newspapers in the United States were already using phrases like “the automobile nuisance.” In France, the catastrophic Paris to Madrid race of 1903 ended with multiple deaths and was halted by authorities before reaching Madrid. Early automobiles did not arrive wrapped in the language of safety or environmental progress. They arrived trailing smoke, blood, noise, and class resentment.

The class issue mattered enormously. Early motorists were mostly wealthy people. To many ordinary citizens, the automobile looked less like transportation and more like aristocratic arrogance mechanized. Woodrow Wilson once remarked that cars generated “socialistic feeling” among rural populations because they symbolized the “arrogance of wealth.” Popular literature absorbed this resentment. In The Wind in the Willows, Mr. Toad becomes intoxicated by motorcars, turning into a ridiculous and dangerous fanatic. In The Magnificent Ambersons, one character dismisses automobiles as “a useless nuisance,” a line that sounded perfectly reasonable to many readers in 1918.

Some places resisted for decades. Mackinac Island banned automobiles in 1898 after carriage operators complained that cars endangered horses and pedestrians. The island remains largely car free today. In the Swiss canton of Graubünden, automobiles were prohibited after residents protested the “speed, noise, and smell” of motor traffic. The ban survived repeated referenda before finally collapsing in 1925. What is striking in retrospect is not that these societies resisted technology. It is that they negotiated with it. They imposed limits, argued about costs, defended existing ways of life, and demanded compensation for disruption.

That is the real lesson of the automobile. The story is not that society always loses when it resists technology. Nor is it a simple morality tale about fearful humans standing against inevitable machines. Many early objections to automobiles were correct. Cars did kill people. They did destroy older industries and reshape cities around themselves. They did transfer public space toward private mobility. What changed over time was the balance of interests. The benefits of automobiles gradually became large enough that more people were willing to tolerate the costs. New industries emerged. Rural isolation diminished. Trade accelerated. Personal mobility expanded. The conflict was never humans versus machines. It was always humans versus other humans, each group attaching different values to speed, safety, convenience, labor, status, and ways of living. Or, in many cases, the same humans against themselves. 

AI Will Write the Lesson Plan, and Teacher Preparation Should Be Glad

Lesson planning is problem solving, and teacher preparation has been treating it as something else. The artifact has been mistaken for the skill. A novice produces a plan that meets the rubric, the supervisor checks the boxes, and everyone agrees the candidate is ready. The plan is the visible output. The reasoning that produced it, which is the actual job, has been left largely to chance and to the quality of mentorship in the placement school. AI is about to make this confusion impossible to sustain, and the news is good rather than bad.

Consider what a teacher actually does when planning a lesson. She holds in mind the standards she must address, the students she has (who they are, what they did yesterday, what they fear, what they already know), the time available, the materials at hand, the colleague next door whose schedule overlaps, the assessment due Friday, the child who was crying in the hall this morning. She makes a sequence of decisions under all these constraints. The plan is the residue of that work. A second teacher with the same standards and a different group of children produces a different plan, and the difference is not stylistic. It reflects different problems being solved.

Programs do not generally teach lesson planning this way. They teach the form. The plan template, the objective-and-anticipatory-set sequence, the alignment to a standard. These are the easy parts, and they are also the parts AI now performs at a competent first-year level. A model handed a standard, a grade, and a topic will produce a workable plan in seconds. If the form was the skill, the skill is gone. If the form was the residue, what was always missing from preparation is what remains, and what remains is the actual teaching job.

I have spent enough years near teacher preparation to know that the field has long understood this in principle. The actual teaching job is solving complex, contextual problems, both pedagogical and relational, repeatedly. The repetition is the part that matters. Judgment is not transmitted by lecture. It accretes through volume of supervised decisions under varied conditions. A candidate should encounter hundreds of small cases before graduating. A reading group is stalling on a text from which two students have visibly disengaged. A child has stopped speaking in class after a family event the teacher heard about indirectly. A planned activity is collapsing in second period and four periods remain. Each case demands a decision. The decision can be defended or critiqued. The next case is different enough that the previous lesson does not simply transfer. This is how master teachers were built when the on-ramp existed in schools. The on-ramp has been narrowing for years. AI does not narrow it further. AI provides the platform on which preparation programs can finally do this work at scale.

Here is the part that should reorganize the curriculum. Students should learn to feed the complexity of a real classroom into the AI, and to evaluate what comes back. Not "write me a lesson plan on the water cycle for fifth grade." That prompt deserves the generic plan it gets. Rather, "I have twenty-eight fifth graders, six of whom read two years below grade, four English learners at different proficiency levels, a classroom with no working sink, forty-five minutes, and a state standard that asks for an investigation. Yesterday's lesson left them confused about evaporation. Propose three different approaches and explain the trade-offs." The skill lives in the input and the evaluation, both of which require knowing what could go wrong with each option. That is judgment in the precise sense, applied through the medium of a competent partner. The AI plans. The teacher decides.

Other professions discovered this structure long ago. Business case method has been the spine of MBA programs for a century, because case after case forces the student to decide under conditions the textbook did not anticipate. Law clinics serve the same function for the willing. Medical residency, at its best, does too. Teacher preparation has the analogous structure in student teaching, but student teaching arrives at the end, depends on the placement, and rarely produces deliberate variation across cases. The model is right. The proportion is wrong.

Putting this into practice means a smaller share of credits on form, methods, and standards memorization, and a much larger share on repeated case decisions with AI as the planning partner. It means faculty time for serious case construction and supervision, which is the most expensive thing a program can spend. It means accepting that some of what currently fills the curriculum is now cheap and must be reallocated rather than retained out of habit. None of this is free, and pretending otherwise will produce a watered-down version that disappoints everyone. The genuine version, however, is something teacher preparation has wanted to be for fifty years and never had the conditions to become. The conditions have arrived.

Slop In, Slop Out

The common way of talking about AI-generated text begins with a category mistake. People want to know what percentage of a piece was written by AI, as though authorship were a divisible substance. Even the more sophisticated framing, that good content is a blend of human and machine work, retains the assumption that the two contributions can be measured against each other on the same scale. They cannot. A prompt of six unusual words can shape an output more decisively than a thousand words of generic instruction. Size is not the right unit, and there is no right unit, because the contributions are not comparable in kind.

The economist David Ricardo described, two centuries ago, a structure that fits this situation better than any ratio. Two parties with different productivity profiles still gain from specialization, even when one is absolutely more capable at every task. The relevant variable is not absolute capability but opportunity cost. England and Portugal both produced wine and cloth in his example, and Portugal was better at both. They still gained by specializing, because the opportunity cost of each good differed between the two countries. Translated into our problem: it does not matter whether AI can technically generate a thesis statement or a first draft. What matters is which side has the lower opportunity cost for which contribution.

The taxonomy is becoming legible. AI systems offer their distinctive value in coverage and convention. They have read more than any individual human can read. They know what a competent paragraph in a given genre looks like, what citations a given claim usually carries, what the statistical regularities of expert prose tend to be. Humans offer their distinctive value elsewhere. The novel connection between two rarely connected fields. The judgment that a particular argument matters and another does not. The intent that determines what is worth writing in the first place. The taste that recognizes when a sentence lands and when it slides past. These are not contributions a model derives from corpus statistics, because corpus statistics describe what has been written, not what should be.

Readers who follow these debates know the term "AI slop," used to describe the bland, formulaic, superficially competent output that floods comment sections and content farms. The term names a result. It does not name the process that produces it. The process deserves its own name. I will call it slop in. The phrase borrows from the old computing maxim that garbage in produces garbage out, and applies it specifically to the moment when a human, sitting in front of a model capable of doing a great deal, supplies an input that wastes the capacity. Slop in is what produces slop out.

The obvious form of slop in is the underspecified request. "Write me an essay about leadership." The human has supplied no intent, no angle, no judgment, and the model fills the vacuum with the average of everything it has seen on leadership, which is the definition of unreadable. But there is a less obvious form, and it is the one I find more interesting. It is the redundant prompt, where the human repeats back to the model what the model would have produced anyway. The user who asks for a five-paragraph essay with a clear thesis, three supporting points, and a conclusion has not contributed what only a human can contribute. They have only specified what the model already defaults to. The output will be worse than if they had said something distinctive, however brief. Effort is not the issue. A long, careful, redundant prompt is still slop in.

J. C. R. Licklider sketched the division of labor we are still working out, in a 1960 essay called "Man-Computer Symbiosis." He drew up two columns. The machine would handle rapid retrieval, pattern matching across volume, and calculation. The human would formulate the questions, choose the criteria of evaluation, and recognize when something was significant. The specifics need updating after sixty-six years, but the structure is exactly right. Licklider understood that the partnership only worked if each side did what the other could not. He did not call it comparative advantage, but the logic is the same.

The pedagogical implication follows. The skill we now need to teach is not how to operate AI tools. The interfaces are not difficult, and the difficult ones will become easy soon enough. The skill is the generation of human input that enhances rather than degrades the joint output. This means learning to recognize what the model will produce by default, so that the contribution can fall elsewhere. It means cultivating the judgment of what is worth writing about, the taste that distinguishes a serviceable sentence from a memorable one, the willingness to assert an interpretation the model would not reach on its own. These have always been the harder parts of writing instruction. They are now also the more economically valuable ones, because they are precisely the inputs the machine cannot supply for itself.

There is a deeper reason this matters, and it has to do with what we are now able to see about ourselves. Every culture defines the human against something it is not. For most of history those somethings were animals and gods, and the contrast did most of the conceptual work. Aristotle's rational animal needed beasts to be irrational. The imago dei needed a creator on the other side of the comparison. The mirrors we used were dim, and the reflection was generous. AI is a new mirror, and a sharper one. It produces fluent argument, competent prose, plausible analysis, and it does so without intent or taste or any judgment of what is worth saying. What remains uniquely human, when the mirror returns this image, is narrower and harder to name than before. Teaching people to contribute to a joint output is therefore not only an economic skill. It is the practical form of the older question about what we are, asked under conditions that finally make the question answerable in detail.