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. 

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.

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.

The Trouble with Refusal Rights: On the CCCC Resolution to Refuse Generative AI

The Conference on College Composition and Communication recently passed a resolution affirming "the rights of students and teachers to refuse to sign up for, prompt, or otherwise use generative AI in the writing classroom." The resolution draws on a companion document, "Refusing GenAI in Writing Studies: A Quickstart Guide" by Jennifer Sano-Franchini, Megan McIntyre, and Maggie Fernandes. I find much of what they say about the political economy of Big Tech persuasive. Yet the central move of casting the argument in the language of rights is a strategic and conceptual misstep.

When we talk about rights, we usually mean one of two things: legal rights, which are enforceable protections backed by law or contract, or moral rights, which are claims about what people are owed regardless of what the law says. The resolution tries to invoke both registers and achieves neither.

The resolution leans on the AAUP's 1940 Statement on Academic Freedom. That document does establish a real professional norm: faculty can select course materials, determine approaches, and assess student work without administrative veto. But notice what this actually protects. It protects faculty autonomy in making curricular decisions. It does not create a specific right to refuse any particular technology. When an instructor decides that a given tool does not serve her course, she exercises normal pedagogical judgment. No one calls that a right. It is simply teaching.

Academic freedom is symmetric. It protects choices, not specific outcomes. A resolution affirming one particular choice as a right tilts the field. It implies that using AI requires justification while refusing it does not.

There is a deeper problem with invoking the AAUP framework here. The AAUP has been explicit that academic freedom in teaching operates on two levels, and the collective level takes precedence. As the AAUP's own FAQ states, the shared academic freedom of a faculty to determine courses and materials "supersedes the freedom of an individual faculty member to choose a textbook that he or she alone prefers." The AAUP's statement on Freedom and Responsibility reinforces this: it is improper for an instructor to fail to present subject matter "as approved by the faculty in their collective responsibility for the curriculum." This is why we have curriculum committees and course approval processes. Individual instructors teach within a collectively sanctioned framework.

The CCCC resolution quietly reverses this logic. It asserts an individual right to refuse a specific technology, independent of any collective deliberation about what a writing curriculum should contain. If we establish the precedent that an individual instructor has a right to refuse AI on principled grounds, what else can an instructor refuse? The learning management system, on grounds that it embodies corporate surveillance? Plagiarism detection software, which the Quickstart Guide itself criticizes? Peer review platforms? Email? Each of these technologies carries ideological baggage. The line between principled refusal and personal preference becomes impossible to draw once you frame the question as one of individual rights rather than collective curricular judgment.

I have also struggled with a simple question while reading the resolution: What specific violation of rights does it aim to prevent? Who is forcing writing teachers to use ChatGPT?

The AAUP's 2025 survey found that 15 percent of faculty said their institution mandates AI use. But the same survey found that 81 percent must use learning management systems with embedded AI features. The "mandate" is mostly that Canvas or Google Workspace now has AI baked in. That is not the same as being told you must assign AI-assisted essays. I cannot find evidence of any American college or university requiring writing faculty to incorporate generative AI into their pedagogy. To the extent that real pressure exists, it takes the form of institutional nudging, not directives that could be resisted by invoking a right.

Rights are most powerful when they address a concrete threat. The right to free speech protects against government censorship. The right to due process protects against arbitrary punishment. What does the right to refuse generative AI protect against? Against being encouraged to try something? Against the zeitgeist? Rights are a heavy instrument. They should be reserved for heavy problems.

Buried inside the resolution is a pedagogical claim: that writing instruction develops human thought and expression, and that outsourcing parts of the writing process to a language model may undermine that development. The claim is debatable but plausible. However, it does not require the language of rights. It requires the language of curriculum and evidence. If generative AI undermines learning outcomes in first-year composition, instructors should not use it. Not because they have a right to refuse, but because it does not serve their students. By framing refusal as a right rather than a pedagogical judgment, the resolution removes the question from the domain where it should be argued and places it in the domain where it can only be asserted. You do not argue against a right. You respect it or you violate it. This forecloses the very inquiry the authors claim to value.

The resolution also affirms a student right to refuse AI. This is even more peculiar. Students are routinely required to use technologies they did not choose: a particular LMS, plagiarism detection services, specific software. No one frames these requirements as rights violations. A course has requirements. An instructor sets them. If the instructor has determined that AI engagement is central to the course, allowing opt-outs means running two parallel courses. If AI is not central, the point is moot.

The real concerns here deserve a better framework. Shared governance matters: if institutions sign contracts with AI companies without consulting faculty, that is a governance problem already addressed by existing norms. Pedagogical autonomy already exists: faculty can and do decide what technologies to use. The evidence question is primary and researchable. And the labor and environmental concerns, while legitimate, are matters of institutional procurement and social policy, not classroom pedagogy. An instructor who refuses ChatGPT on environmental grounds should, by the same logic, refuse Zoom, Canvas, and university email, all of which depend on data centers with significant environmental footprints.

There is a final paradox. The language of rights is supposed to project strength, but here it signals the opposite. When a profession asks for special protections against a new technology, it announces that it cannot figure out, through its own expertise and collective deliberation, how to respond to a changed environment. Math departments did not need a right to refuse calculators. They debated, experimented, and made curricular decisions. Some banned calculators from exams, some required them, and the question was always pedagogical: Does this tool help students learn, or does it let them bypass the thinking? Writing studies has every intellectual resource it needs to conduct the same kind of inquiry. Framing the matter as a right suggests otherwise. It suggests a profession that feels so besieged it must retreat behind quasi-legal barricades rather than do what professions do: deliberate, adapt, and lead.

Why Does AI Feel Like Freedom To Me, Not A Threat To Learning?

I realized early that I am one of those people who love working with AI. The reason is simple: I have always hated routine. For years in higher education administration I tried to automate any small piece of work that I could. I was the person who wrote macros and ran Mail Merge in Word when most colleagues did not know the feature existed (many still don't). I liked thinking about problems and working with people. I did not like documenting those problems afterward.

Administrative work in universities is full of documents that must look serious and official. Strategic plans, accreditation reports, assessment summaries, memos, program reviews. These texts are often written in a careful, dry tone that tries to sound objective and important. For me those documents felt like a toll road that I had to pay to get to the interesting parts of the job. I wanted to talk with faculty, students, and community partners about real issues. I wanted to design new programs and rethink old ones. Instead I spent hours polishing language that almost nobody planned to read.

When large language models became usable, something in me clicked. Within days I was intellectually convinced that this technology would matter for education. But I also had a powerful emotional reaction. This felt like a long delayed form of liberation. A big piece of my working life that had always felt wasted could now be reclaimed for thinking and for human interaction.

I do not experience AI as a threat to meaning or to craftsmanship. I experience it as an assistant that removes the crust from the work. I still need to decide what the document should say, who it is for, what matters and what does not. But I do not need to fight with first drafts, with standard phrases, or with the sheer volume of institutional writing. AI gives me more time for the activities I actually value: creative thinking, brainstorming, problem solving, building theory, and talking with real people.

This personal relief also made me more cautious about judging other people's reactions. Our field often talks about AI as if there is a single rational stance to take. In reality, psychotypes matter. Some people are wired to enjoy the very things that I dislike. They take pleasure in the slow craft of sentence building. They value the aesthetics of a beautiful paragraph, the elegance of a careful transition, the feeling of a page that has no awkward phrase anywhere. The process itself is rewarding for them, not just the outcome. These are not shallow preferences. They reflect different theories of what work should feel like and what makes it meaningful.

There are also people for whom accuracy and authority are core values. They want information to be correct, checked, and stable. They trust texts that feel final. For them, any minor error, even a typo, can cast doubt on the whole product. When they look at AI, they see a tool that produces fluent but sometimes wrong text, and that feels deeply unsafe. The idea that something could sound confident and still be mistaken violates their sense of how knowledge should be handled. Their resistance grows from a coherent set of commitments about what scholarship requires.

My preferences run in a different direction. I care more about speed, access, and the flow of ideas than about perfect reliability at the sentence level. I was an early fan of Wikipedia for exactly this reason. I liked the fact that I could reach a reasonable overview of almost any topic in seconds, even if I knew that some entries had gaps or errors. Many colleagues still treat Wikipedia as second rate. For me, being mostly correct is good enough, as long as I can see interesting ideas and follow references further if needed. This is a different epistemology, not a careless one.

AI feels like an extension of that tradeoff. It gives me fast, flexible text that I can shape, question, and rebuild. I do not expect it to be right in every detail. I expect it to help me think faster and wider. What I cannot easily get without AI is a steady partner that never gets tired of drafting, revising, or trying a different structure for the tenth time. The machine does not care how many times I change my mind. That patience has real value.

In one of my earlier reflections I argued that doing a task very well does not prove that the task itself is worthwhile. AI has pushed that point closer to home. Many academic and administrative texts are produced with great skill, but the value of that effort is not always clear. If a machine can now produce a comparable draft in seconds, it becomes easier to ask what exactly we are adding with our human labor, and whether we want to spend our limited time there. This is an uncomfortable question for professions that have built identity around textual competence.

The issue goes beyond individual preference. Different psychotypes produce different institutional cultures. Organizations dominated by people who value routine and formal documentation will resist AI more strongly than organizations where improvisation and speed are prized. These cultural differences shape what counts as quality, what gets rewarded, and who advances. When AI enters the picture, it does not just change tools. It shifts the balance of power between competing visions of professional life.

When we think about AI in education, we should factor in these differences in temperament. Policy debates tend to focus on abstract risks and benefits. Underneath those arguments lie different ways of relating to work, to text, and to uncertainty. People who experience AI as freedom will advocate for rapid adoption and experimentation. People who experience it as erosion of craft or trust will ask for limits and safeguards. Both groups have a piece of the truth. The challenge is to build systems flexible enough to accommodate both without forcing everyone into the same mold.

Any serious conversation about AI in education has to make space for multiple stories and for the many shades in between. What we cannot afford is to pretend that these differences are merely technical, or that one good argument will settle the matter for everyone. The stakes are partly emotional, partly philosophical, and deeply tied to how we understand the purpose of our work.

Beating the Robot Is the Point (and the Pedagogy)

A pivotal moment in any course involving artificial intelligence comes when students try, and succeed, in beating the robot. I do not mean cheating the system or outsmarting the instructor. I mean learning how to identify something AI does poorly, and then doing it better.

Many students, especially undergraduates, approach AI with exaggerated reverence. They assume the output is authoritative and final. AI writes with confidence, speed, and often impressive fluency. The effect is almost hypnotic. This creates a psychological barrier: if the machine does it well, what is left for me to do? Am I smart enough to compete? 

This assumption is wrong, but not irrational. It takes cognitive effort to move beyond awe toward critique. The breakthrough moment occurs when a student notices a flaw. Sometimes it is a factual error, but more often it is a subtle lack; an absence of argument, weak nuance, robotic phrasing, or flat tone. Students realize, for the first time, that the AI is not a better version of themselves. It is something different. It is stronger in language processing but weaker in creativity, authenticity, judgment, insight, or affect.

This realization is not theoretical. It is a variant of self-efficacy, but more specific and applied. Classic self-efficacy theory describes the conviction that one is capable of performing a task. What occurs in the classroom with AI is more nuanced. Students do not just believe they can do something. They discover what, exactly, they can do better than the machine. This is a kind of enhanced self-efficacy; focused not on general ability, but on identifying one's own unique niche of competence. It is confidence through contrast.

To beat the robot, one must first learn to challenge it. That could mean prompting it more cleverly, iterating multiple drafts, or simply refusing to accept its first answer. Students begin to demand more. They ask, “What is missing here?” or “Can this be said better?” The AI becomes a foil, not a teacher. That shift is vital.

There are students who reach this point quickly. They treat AI as a flawed collaborator and instinctively wrestle with its output. But many do not. For them, scaffolding is necessary. They must be taught how to critique the machine. They must be shown examples of mediocre AI-generated work and invited to improve it. This is not a lesson about ethics or plagiarism. It is a lesson about confidence.

Cognitive load helps explain why some students freeze in front of AI. The interface appears simple, but the mental task is complex: reading critically (and AI is often verbose), prompting strategically, evaluating output, and iterating; all while managing anxiety about technology. The extraneous load is high, especially for those who are not fluent writers. But once a student identifies one specific area, such as tone, logic, detail, where they outperform the machine, they begin to reclaim agency. That is the learning goal. I sometimes explain it to them as moving from the passenger seat to the driver's seat. 

This is not the death of authorship. It is its rebirth under new conditions. Authorship now includes orchestration: deciding when and how to use AI, and how to push past its limitations. This is a higher-order skill. It resembles conducting, not composing. But the cognitive work is no less real.

Educators must design for this. Assignments should not simply allow AI use; they should require it. But more importantly, they should require critique of AI. Where is it wrong? Where is it boring? Where does it miss the point? Students should be evaluated not on how well they mimic AI but on how well they improve it.

Some students will resist. A few may never get there. But most will, especially if we frame the challenge not as compliance, but as competition. Beating the robot is possible. In fact, it is the point. It is how students learn to see themselves not as users of tools but as thinkers with judgment. The robot is fast, but it is not wise. That is where we come in.

Educating for a Simulated Relationship

As AI settles into classrooms, we face a peculiar challenge: not just how students use it, but how they relate to it. This isn’t a question of function or ethics, but of posture—how to engage something that responds like a person but isn’t one. The educational goal is a subtle kind of literacy: to treat AI as an interactive partner without mistaking it for a peer.

It’s a familiar dilemma, strangely enough. When children talk to imaginary friends or fictional characters, they often treat them as real companions. They know, on some level, that the character isn’t real—but they still cry when the story ends or feel comforted by a plush animal’s imagined voice. Child psychologists don’t rush to correct this confusion. Instead, they guide children to inhabit the fiction while understanding its boundaries. The fiction is developmental—it helps the child grow, not deceive.

We need a similar stance with AI. Students must learn to engage in what we might call a non-dialogic dialogue: a back-and-forth that mimics human exchange but is, in substance, interaction with an “It.” Martin Buber’s language is useful here. Procedurally, AI feels like an “I-Thou”—responsive, adaptive, present. But substantively, it remains an “I-It.” It has no inner life, no perspective, no sense of being addressed.

If we treat AI merely as a tool, we lose its pedagogical richness. If we treat it as a mind, we delude ourselves. The path forward is both instrumental and interactive: act as if the AI understands, but always know it doesn’t. This requires a new kind of mental discipline—AI mind theory, if you like. Not to imagine what AI thinks, but to restrain the impulse to imagine that it does at all.

In practice, this means teaching students to hold contradiction. To benefit from AI’s apparent collaboration, without anthropomorphizing it. To take seriously its output, without confusing fluency with insight. It’s a balancing act, but one education is well suited for. After all, school isn’t meant to tidy up complexity. It’s meant to make us capable of thinking in layers.

AI is not our friend, not our enemy, not even our colleague. It is something stranger: a fiction we interact with for real.

Executive Cognition, a New Learning Outcome We Cannot Ignore

Education has always been, at its core, a wager on the future. It prepares students not only for the world that is, but for the world that might be. Yet increasingly, the curricula of our universities and schools are anchored to a past that no longer exists. They reward intellectual independence in a time that demands cognitive interdependence. They emphasize content mastery while overlooking the far more urgent task of managing cognition itself—particularly when that cognition is now distributed across humans and machines.

The rise of artificial intelligence has not simply added another tool to the professional toolbox. It has fundamentally restructured how intellectual work is done. Tasks that once defined expertise—drafting legal memos, analyzing medical data, composing arguments—can now be executed by algorithms. The value of education is no longer found in preparing students to do all the thinking themselves. It lies, instead, in preparing them to orchestrate thinking across human and non-human systems. This is not an incremental shift. It is a paradigmatic one.

What this moment demands is a new kind of learning outcome. Not technical fluency alone. Not ethics in abstraction. But a metacognitive capacity to coordinate, delegate, and evaluate in environments where machine intelligence plays a central role. This capacity, best described as executive cognition, is neither a buzzword nor a future aspiration. It is the most neglected, yet most necessary, educational objective of our time.

Executive cognition refers to the ability to strategically allocate attention, manage cognitive resources, and construct workflows that integrate AI systems without surrendering human judgment. It is what separates those who can simply operate tools from those who can govern systems of intelligence. It requires building accurate mental models of what AI can and cannot do—not from a technical standpoint, but from a cognitive one. It demands recognition that AI can generate but not reason, correlate but not explain, perform but not understand.

These are not abstract distinctions. They play out daily across professions. In law, attorneys now spend less time writing and more time evaluating AI-generated arguments. In medicine, physicians must decide when to trust diagnostic algorithms and when to intervene. In business, executives orchestrate AI analyses but retain accountability for their implications. The work has not disappeared; it has moved. And education must follow it.

The problem is that it has not. Most institutions continue to assess students on their ability to produce knowledge alone. The quality of the final product remains the measure of success, even as in professional life the process—how one decides what to delegate, how one interprets outputs, how one integrates disparate sources—has become the true mark of expertise. Students are seldom taught how to think with machines, only how to think despite them.

This inertia is not just pedagogical. It is philosophical. It reflects a stubborn belief that the highest form of learning is independent intellectual production. But that ideal is now, in practice, a fiction. No serious professional today operates without digital augmentation. The challenge is no longer to preserve human reasoning by avoiding AI, but to elevate human reasoning by directing AI.

There is, of course, a danger in teaching students to rely on systems they do not fully understand. But there is a far greater danger in failing to teach them how to rely wisely. Without executive cognition, students will either misuse AI or abdicate responsibility to it. Neither approach fosters the judgment we claim to value.

To center executive cognition in education is not to diminish traditional skills. It is to recontextualize them. Writing, analysis, and interpretation remain essential—but they are now embedded within larger cognitive ecosystems. Knowing how to frame a question is just as important as knowing how to answer it. Knowing when to withhold trust in an algorithm is more valuable than accepting its efficiency uncritically. The future belongs not to the fastest thinkers, but to the most discerning orchestrators.

Some will object that education should not adjust so quickly to technological change. That we risk becoming vocational rather than intellectual. But the cultivation of executive cognition is not a capitulation to industry; it is a reaffirmation of education’s highest purpose: to develop judgment. It merely recognizes that the forms of judgment required today are different from those of the past.

What we need now is not more information, but better navigation. Not more tools, but wiser tool use. Executive cognition provides a conceptual and practical framework for this shift. It does not ask educators to become engineers, but to become curators of thinking—guiding students not toward mastery of content, but toward mastery of cognitive delegation, integration, and oversight.

This is not the only learning outcome that matters, but it may be the most pressing one we have neglected. It is time for education to make a new wager—one that aligns more closely with the world students will inherit than with the one their educators were trained to navigate. The machine can generate. Only the human can judge. But only if we teach them how.

Why Education Clings to Irrelevance

There’s a particular irony in the way schools prepare us for life by arming us with tools designed for a world that no longer exists. Latin, long dead yet stubbornly alive in syllabi. Algebra, elegantly abstract but only marginally helpful in daily decisions compared to statistics, which might actually help you not get swindled by a mortgage. Even now, when artificial intelligence knocks at the schoolhouse door, the impulse is to double down on handwriting and closed-book tests, as if graphite and paranoia can hold off the machine age.

The problem isn’t just inertia. It’s something deeper—almost spiritual. Education doesn’t merely lag behind practicality. It seems to celebrate its irrelevance. Irrelevance isn’t a bug of the system. It’s the feature. Like a stubborn elder at the dinner table insisting that things were better “back in the day,” the institution of education resists change not just because it can, but because its authority is rooted in not changing.

To teach is, implicitly, to claim authority. But where does this authority come from? Not the future—that’s terra incognita. You can't be an expert in what hasn't happened yet. So the educator turns to the past, because the past has already been canonized. It has textbooks. It has footnotes. It has certainty. And nothing flatters the authority of a teacher more than certainty. Thus, to preserve this role, education must tie itself to a vision of life that is already archived.

There’s a paradox here: education is supposedly preparation for life, yet it often refuses to adapt to the life that is actually unfolding. It is conservative in the truest sense—not politically, but ontologically. It conserves the past. It prepares students for a utopia that once was, or more precisely, that someone once imagined was. The classroom becomes a time capsule, not a launchpad.

This is not entirely wrongheaded. The future is uncertain, and while we might guess at the skills it will demand—data literacy, adaptability, collaborative creativity—we cannot guarantee them. So we retreat to what we can guarantee: the conjugation of Latin verbs, the proof of triangles, the essays written in blue ink. It's a strategy of safety cloaked as rigor. If we can’t foresee what will matter tomorrow, at least we can be very confident about what mattered yesterday.

And yet, this nostalgia has consequences. It breeds irrelevance not just in content, but in spirit. When students ask, “When will I ever use this?” and teachers respond with some forced scenario involving imaginary apples or train schedules, the real answer—“you probably won’t, but we’re doing it anyway”—lurks just beneath the surface. The curriculum becomes a kind of ritual, an educational incense burned in memory of older truths.

The arrival of AI only sharpens this tension. Faced with the destabilizing presence of machines that can write, summarize, solve, and simulate, many educators panic. But rather than adapting, they retreat further. The pencil becomes a moral statement. The closed classroom a sanctuary. There’s almost a theological quality to it, as if real learning must involve a kind of suffering, and anything too efficient is suspect.

It’s tempting to dismiss this all as folly, but maybe it reflects something deeply human. Our fear of irrelevance leads us to preserve what is already irrelevant, in hopes that it might make us relevant again. In this way, education mirrors its creators: creatures of habit, haunted by the past, nervous about the future, and always unsure where real authority should lie.

Perhaps that’s the real lesson schools teach, intentionally or not. That relevance is fragile, and that meaning is something we inherit before we can create it. 

The Invasive Innovation, a Case of AI in Education

Disruptive innovations, as defined by Clayton Christensen, typically emerge at the margins before gradually overtaking established players. In contrast, artificial intelligence in education is not a slow-moving force of competitive displacement. It is an invasive innovation—spreading far beyond the predictions of Everett Rogers’ diffusion model, propelled not as much by institutions as by students themselves. This is not a technology waiting to be adopted. It is a reality forcing adaptation. Treating it as an optional innovation mises the point entirely. 

Educational leaders, policymakers, and institutions may wish to evaluate AI’s impact, consider best practices, and implement careful frameworks. But they are not in control of the timeline. The adoption is not up for debate. It is already happening, driven by students who see AI not as a tool of the future but as an everyday utility, as natural as search engines or calculators. The question is not when to integrate AI into education, but how to manage an environment where AI is already integrated—often in ways institutions neither designed nor anticipated.

This is why traditional approaches to educational innovations are inadequate. Normally, innovations are assessed, piloted, and then scaled. But AI in education is unfolding more like a crisis than an initiative. It requires a response more akin to a public health emergency than a strategic rollout. When COVID-19 disrupted education, schools did not spend years analyzing whether remote learning was effective. They pivoted overnight, creating imperfect but necessary solutions. AI invasion demands a similar level of urgency.

Institutional response to invasive AI must be comprehensive rather than defensive. How do we accommodate this invasive innovation? Putting AI tools directly in the hands of faculty represents a critical first step, ensuring they engage with the same technologies their students already access freely. This must be paired with robust professional development—not merely technical training but genuine pedagogical exploration that helps educators reimagine their disciplinary approaches in an AI-augmented world.

Beyond tools and training, institutions must invest in creating dedicated time and space for disciplinary teams to collaboratively revise their assignments. The goal is not to create "cheat-proof" assessments but to develop authentic tasks that meaningfully incorporate AI as a partner in learning. This demands protected time away from everyday teaching pressures, where faculty can experiment, fail, refine, and document new approaches that acknowledge AI as a permanent fixture in their educational landscape rather than an external threat to be neutralized.

Sidorkin Hypothesis: Beyond Procedural Fluency in AI-Augmented Instruction

The hypothesis is this: Quality AI-augmented instruction reduces emphasis on procedural skills while enhancing higher-order thinking and conceptual learning. This shift may offer an alternative pathway for cognitive offloading, which could supplement or even replace traditional procedural skills acquisition.

Cognitive load theory, developed by John Sweller in the 1980s, provides a useful framework. The theory posits that our working memory has limited capacity when dealing with new information. Sweller himself, along with many other cognitive scientists, views fluency in procedural skills as a major mechanism for cognitive offloading. When basic procedures become automatic through practice, they consume fewer working memory resources, thereby freeing up mental capacity for higher-order learning. This is why traditional education emphasizes mastering procedural skills—calculating derivatives, balancing chemical equations, applying grammatical rules—before tackling complex conceptual work.

In my view, AI tools function as an alternative cognitive offloading mechanism that can complement or even supersede traditional procedural fluency. These tools handle procedural tasks , creating cognitive space for students to engage with concepts at a deeper level without first developing procedural automaticity. Consider a student learning calculus: traditionally, substantial mental resources are dedicated to memorizing differentiation rules and practicing their application. With AI assistance handling these procedural aspects, the student can direct cognitive resources toward understanding conceptual foundations—what derivatives actually represent, how they relate to real-world phenomena, and when different approaches should be applied.

This approach necessitates reimagining assessment. Traditional writing instruction heavily rewards procedural accuracy in grammar, citation formats, and structural conventions, but in an AI-augmented educational landscape, assessment must evolve to measure the ability to prompt AI tools effectively, the quality of critical analysis, application of rhetorical strategies in novel contexts, and evaluation of AI-generated text. The focus shifts from "Can you properly format this essay?" to "Can you determine which rhetorical approach best suits your purpose, interpret complex texts critically, and evaluate whether the AI-generated analysis captures the deeper meaning of the work being examined?"

This transformation faces substantial challenges. Educators rightfully question whether students can truly grasp concepts without some level of procedural fluency. it is likely that some basic  procedural knowledge remains necessary for conceptual understanding. The optimal approach involves thoughtful integration—using AI to reduce procedural load while ensuring students develop core competencies. To put it roughly, "procedural knowledge - yes, procedural fluency - not, really."  

The equalizing potential of AI-augmented instruction cannot be overlooked. Students who traditionally struggle with procedural aspects—whether due to learning differences, educational disruptions, or lack of prior opportunities—may find themselves able to participate more fully in higher-order learning. This democratization of advanced conceptual engagement could elevate educational outcomes across diverse student populations. We stand at the threshold of potentially transforming education from a system that sorts students based on procedural proficiency to one that cultivates universally richer, more advanced learning outcomes: deeper critical analysis, more sophisticated problem formulation, nuanced ethical reasoning, and creative synthesis across domains—intellectual capabilities that more authentically reflect the complex demands of contemporary society.

Form-substance discrimination, a new learning outcome

We have long assumed that clear writing signals clear thinking. Schools teach writing as if it were math - a rigorous exercise that develops logical thinking. Editors catch not just errors but muddled ideas. Or so the theory goes. Artificial intelligence shatters this comfortable assumption by churning out impeccably structured prose devoid of original thought.

Form-substance discrimination represents a higher-order cognitive skill, similar to what art historians develop when they learn to separate technical mastery from creative vision. Just as an art student must unlearn their initial attraction to photorealistic paintings, readers now must develop resistance to seductively clear prose. This requires a kind of cognitive inhibition - the ability to suppress immediate aesthetic pleasure for the sake of deeper analysis.

The skill builds on existing metacognitive abilities but requires their novel application. Readers already know how to identify main ideas, analyze arguments, and evaluate evidence. What is new is the need to perform these operations while actively discounting the surface appeal of the text. This resembles what wine tasters do when they evaluate wines blind, stripped of prestigious labels and beautiful bottles.

The development follows a predictable pattern. At first, readers struggle to overcome their ingrained respect for well-crafted prose. The initial challenge lies not in identifying weak ideas but in giving oneself permission to criticize a text that follows all the rules of good writing. This mirrors the development of critical thinking in general, where students must learn to question authority figures who appear competent and confident.

The second stage involves developing specific techniques for idea extraction. Readers learn to create idea maps independent of the text's structure, to count unique concepts rather than words, to identify circular arguments hidden behind elegant transitions. They begin to see how AI-generated text often creates an illusion of logical flow while merely restating the same point in different words.

The final stage brings automaticity. Experienced readers develop an immediate sense of a text's intellectual weight, just as experienced teachers can quickly gauge a student's understanding despite fluent recitation. This involves pattern recognition built through exposure to many examples of both substantive and hollow texts.

The educational implications are significant. Writing instruction must now explicitly separate craft from content. Students need exposure to both well-written texts with weak ideas and poorly written texts with strong ideas. They must practice identifying when sophisticated language masks conceptual poverty and when rough expression contains genuine insight.

This shift parallels broader changes in how we process information. In a world of information abundance, the key skill is no longer finding or producing well-formed content but discerning its value. Form-substance discrimination represents a specific case of this general challenge - learning to navigate a world where traditional quality signals no longer reliably indicate underlying worth.

The skill matters beyond academia. Business leaders reading AI-generated reports, citizens evaluating political arguments, professionals studying their field's literature - all need to separate rhetorical sophistication from intellectual contribution. As AI writing tools improve, this ability will become as fundamental as basic literacy.

We face a paradox: the better AI becomes at writing, the more important it becomes for humans to see through good writing. The very perfection of AI-generated prose creates a new kind of opacity that readers must learn to penetrate. Form-substance discrimination thus emerges not just as an academic skill but as a key component of modern critical thinking.

Is Critical Thinking Going Extinct? Maybe That's Not Bad

As someone who remembers using paper maps and phone books, I find myself fascinated by Michael Gerlich's new study in Societies about AI's impact on our cognitive skills. Those of us who learned to navigate by landmarks and memorized phone numbers often bemoan younger generations' reliance on digital tools. But perhaps we are missing something important about cognitive evolution.

Gerlich's research is methodologically elegant. Through surveys and interviews with 666 participants, he documents a decline in traditional critical thinking skills among frequent AI users. The data analysis is rigorous - multiple regression, ANOVA, random forest regression - showing clear correlations between AI tool usage and reduced traditional analytical thinking.

But here's where I think Gerlich misses a crucial insight. The study measures critical thinking through metrics developed for a pre-AI world. It's like judging modern urban survival skills by the standards of hunter-gatherer societies. Those ancient peoples could track game, identify countless plants, and navigate vast territories without maps. By their standards, most of us would be considered cognitively impaired.

What we're witnessing is not cognitive decline but cognitive adaptation. Today's "critical thinking" is not about solving problems independently - it's about effective human-AI collaboration. It's about knowing when to trust AI and when to question it, how to frame queries effectively, and how to combine AI insights with human judgment.

The educational implications are profound. Instead of lamenting the loss of traditional cognitive skills, we should focus on developing "AI-literate critical thinking." Sure, I can still read a map, but my children need to master skills I never dreamed of - like crafting effective prompts for AI systems or critically evaluating AI-generated content.

The old form of critical thinking might be fading, like the ability to start a fire by friction or navigate by stars. But a new form is emerging, better suited to our technological reality. Our task is not to resist this evolution but to guide it wisely.

What do you think? Are we really losing something irreplaceable, or are we just adapting to a new cognitive environment?

Where is the work? AI and Creativity

For ages, we have blurred the lines between ideation and execution, treating them as inseparable parts of creativity. Craftsmanship was tightly bound to originality. Think of Michelangelo working on the Sistine Chapel, a project that spanned nearly a decade. Where does his genius truly lie? In envisioning those profound images, or in the labor of painting them? What, exactly, is the essence of the work?

The rise of AI forces us to untangle these ideas and reconsider what it means to produce "human" work. Take a recent story I heard from from the audience of one of my talks: a person described how he fed an AI every detail about a retiring colleague, and the AI generated a speech so moving that it brought the retiree to tears. But the retiree, upon learning the speech's origin, was dumbfounded.

What is interesting is not the retiree’s reaction, but the storyteller's own oversight. He failed to see his own critical role in the process. By gathering the details, curating moments that best captured the retiree’s essence, he performed the most human part of the creative act. He mistook the act of turning those ideas into words as the creative work, but that is not the case.

AI, ironically, is pushing us to be more human, not more like machines. It is forcing us to recognize that our true contribution lies in the ability to think, to create, and to feel. As AI takes over the mechanical aspects of tasks we once considered integral to creativity—whether that is writing, painting, or coding—we are left with the more uniquely human roles: original thinking and emotional depth.

This shift reshapes our understanding of creativity and work. It shows that human value does not lie in production—the technical aspect of turning an idea into a product—but in the deeper conceptual and emotional layers that AI still cannot reach.

As we move forward, we are compelled to rethink productivity itself. The future will not belong to those who can outdo AI in execution, but to those who can combine AI’s strengths with our unique capacities for innovation, empathy, and insight.

The challenge we face is not to resist AI, but to fully embrace our humanity—to cultivate the traits that machines cannot replicate. With AI taking over the drudgery, we are freed to focus on higher-order thinking and those creative leaps that define human ingenuity.

Ironically, the more we develop artificial intelligence, the more we learn about what human intelligence really is. And in that discovery lies our future—a future where AI does not replace creativity, but elevates it to new possibilities.

Advanced AI users develop special cognitive models

When we encounter a stranger, we make swift, often unconscious judgments about who they are and what they are capable of. A person who speaks our language with barely a hint of an accent? We assume they are fluent. Someone who drops a reference to a complex scientific theory? We peg them as well-educated, likely to be literate, and probably knowledgeable about a range of topics from current events to social norms.

These snap judgments form the backbone of our social interactions. They are mental shortcuts, honed over millennia of human evolution, allowing us to navigate the complexities of social life with remarkable efficiency. Most of the time, they serve us well. We can usually guess whether someone will understand a joke, follow a complex argument, or need help using a smartphone. These are cognitive models. 

But when we step into the realm of artificial intelligence, these time-tested models crumble. Our human-centric predictions fail spectacularly, leaving us confused and often frustrated. Consider a recent incident with ChatGPT, a sophisticated language model. When asked to count the number of 'r's in the word "strawberry," it faltered. Many observers scoffed, concluding that AI must be fundamentally stupid if it couldn't handle such a simple task.

Yet this reaction reveals more about our flawed expectations than any shortcoming of AI. Those familiar with AI's inner workings were not surprised. They understand that a language model, no matter how advanced, is not optimized for character-level analysis. It is like expecting a master chef to be an expert accountant simply because both professions involve numbers.

This misalignment between our expectations and AI's actual capabilities stems from our tendency to anthropomorphize. We instinctively attribute human-like qualities to these digital entities. We expect them to have consistent opinions, to learn from our interactions, to understand context and nuance as we do. But AI, in its current form, does none of these things.

Unlike humans, AI does not carry the baggage of personal experience or emotion. It does not have good days or bad days. It will not be flattered by praise or offended by insults. It can switch from discussing quantum physics to writing poetry without missing a beat, unencumbered by the specialization that defines human expertise.

But AI's differences extend beyond mere capability. It lacks the fundamental attributes we associate with consciousness. It has no self-awareness, no goals or motivations of its own. It does not truly understand the content it generates, despite how convincing it may seem. It is a reflection of the data it was trained on, not a sentient being forming its own thoughts and opinions.

To interact effectively with AI, we need to develop new mental models. We must learn to predict its behavior not based on human analogies, but on an understanding of its unique nature. This means recognizing that AI might struggle with tasks we find trivially easy, while effortlessly accomplishing feats that would challenge even the most brilliant human minds.

It means understanding that every interaction with AI is essentially new. Unlike humans, who build on past conversations and experiences, most current AI systems do not retain information from one chat to the next. They do not learn or evolve through our interactions. Each query is processed afresh, without the context of what came before.

This new model of understanding also requires us to be more precise in our interactions with AI. While humans often fill in gaps in conversation with assumed context, AI interprets our requests literally. It does not automatically infer our unstated needs or desires. The clarity of our input directly influences the quality of the AI's output.

As AI becomes an increasingly integral part of our lives, developing these new mental models is not just about avoiding frustration. It is about unlocking the full potential of these powerful tools. By understanding AI's strengths and limitations, we can craft our interactions to leverage its capabilities more effectively.

The future of human-AI interaction lies not in expecting AI to conform to human patterns, but in adapting our approach to align with AI's unique characteristics. It is a future that requires us to be more thoughtful, more precise, and more open to rethinking our instinctive assumptions. In doing so, we may not only improve our interactions with AI but also gain new insights into the nature of intelligence itself. 

Cognitive Offloading: Learning more by doing less

In the AI-rich environment, educators and learners alike are grappling with a seeming paradox: how can we enhance cognitive growth by doing less? The answer lies in the concept of cognitive offloading, a phenomenon that is gaining increasing attention in cognitive science and educational circles.

Cognitive offloading, as defined by Risko and Gilbert (2016) in their seminal paper "Cognitive Offloading," is "the use of physical action to alter the information processing requirements of a task so as to reduce cognitive demand." In other words, it is about leveraging external tools and resources to ease the mental burden of cognitive tasks.

Some educators mistakenly believe that any cognitive effort is beneficial for growth and development. However, this perspective overlooks the crucial role of cognitive offloading in effective learning. As Risko and Gilbert point out, "Offloading cognition helps us to overcome such capacity limitations, minimize computational effort, and achieve cognitive feats that would not otherwise be possible."

The ability to effectively offload cognitive tasks has always been important for human cognition. Throughout history, we've developed tools and strategies to extend our mental capabilities, from simple note-taking to complex computational devices. However, the advent of AI has made this skill more crucial than ever before.

With AI, we are not just offloading simple calculations or memory tasks; we are potentially shifting complex analytical and creative processes to these powerful tools. This new landscape requires a sophisticated understanding of AI capabilities and limitations. More importantly, it demands the ability to strategically split tasks into elements that can be offloaded to AI and those that require human cognition.

This skill - the ability to effectively partition cognitive tasks between human and AI - is becoming a key challenge for contemporary pedagogy. It is not just about using AI as a tool, but about understanding how to integrate AI into our cognitive processes in a way that enhances rather than replaces human thinking.

As Risko and Gilbert note, "the propensity to offload cognition is influenced by the internal cognitive demands that would otherwise be necessary." In the context of AI, this means learners need to develop a nuanced understanding of when AI can reduce cognitive load in beneficial ways, and when human cognition is irreplaceable.

For educators, this presents both a challenge and an opportunity. The challenge lies in teaching students not just how to use AI tools, but how to think about using them. This involves developing metacognitive skills that allow students to analyze tasks, assess AI capabilities, and make strategic decisions about cognitive offloading.

The opportunity, however, is immense. By embracing cognitive offloading and teaching students how to effectively leverage AI, we can potentially unlock new levels of human cognitive performance. We are not just making learning easier; we are expanding the boundaries of what is learnable.

It is crucial to recognize the value of cognitive offloading and develop sophisticated strategies for its use. The paradox of doing less to learn more is not just a quirk of our technological age; it is a key to unlocking human potential in a world of ever-increasing complexity. The true measure of intelligence in the AI era may well be the ability to know when to think for ourselves, and when to let AI do the thinking for us.