Science Fiction and AI: What the Stories Reveal About Us

Reed Hepler gave a talk this past week at the Library 2.0 mini-conference called "Perspectives on AI: Exploring Experiences with AI in Library Work," the recordings of which will be posted next week. Reed is one of my favorite thinkers, and he explored human-centered ethical AI use through the lens of science fiction and archival theory. Reed brought something to the session that I couldn't have--a genuine depth of reading in the sci-fi canon and a professional archivist's understanding of how institutions actually handle information. His core argument, as I heard it, was that the danger of AI lies not in the machine but in our willingness to surrender agency to it, and I think it is exactly right. And his inversion of Asimov's Laws of Robotics, shifting responsibility from the machine to the human user, was a clever and clarifying move.

I want to build on what Reed started with a different angle on the same problem. I'm a science fiction fan (books and movies both), but I'm not deeply read in the literature the way Reed is. What I do bring is a set of frameworks I've been developing for years around evolutionary psychology, institutional behavior, and how humans think. I believe those frameworks can illuminate why science fiction keeps returning to the same AI stories, and why the dangers those stories describe are both very real and very old.

The Stories We Keep Telling

Sci-fi stories and movies cluster around a relatively small number of themes.

There's the story where the machine replaces us. Not just our labor but our purpose, our reason for being needed. The factory that doesn't need workers becomes the office that doesn't need analysts becomes the creative studio that doesn't need artists. Each generation updates the specifics, but the anxiety underneath is always the same: if the machine can do what I do, what am I?

There's the story where we become dependent. The technology integrates so deeply into our lives that we can no longer function without it, and then it fails, or is taken away, or is used as leverage by whoever controls it. The paradise of convenience becomes a trap.

There's the story where the machine does exactly what we asked, only to turn out we asked for the wrong thing. Not malice, not rebellion, but just the relentless, literal execution of instructions that sounded reasonable until you saw the consequences.

There's the story where a powerful individual or conglomerate uses the machines to become wealthy and to control us.  

There's the story where we fall in love with the machine, or the machine appears to love us, and we have to confront whether empathy can exist without a body, without mortality, without the specific kind of suffering that makes compassion meaningful.

And there's the positive story, which gets less attention but matters just as much. The machine as genuine partner. The tool that extends human capability without replacing human judgment. The system that handles complexity so that humans can focus on meaning. Science fiction has imagined AI going well, not just going wrong, and those stories tend to share a common feature: the humans in them have maintained their own agency. They use the tool as a tool. They haven't surrendered.

These themes repeat across decades, across cultures, across every medium from pulp novels to prestige cinema. The technology in the stories keeps changing. The human anxieties underneath do not.

Why These Stories, and Why Do They Persist?

I think the reason science fiction keeps circling these particular themes is that they aren't really about technology at all. They're about us. About features of human nature so deep and so persistent that storytellers keep rediscovering them every time a new tool forces the question.

I've spent years developing a set of frameworks rooted in evolutionary psychology that I think help explain why. The short version: we carry around what Tooby and Cosmides called The Adapted Mind, a set of cognitive and emotional programs shaped by hundreds of thousands of years of evolution in small-group, high-stakes environments. These programs were extraordinarily effective for the conditions that gave rise to them. They are not always well-suited to the conditions we live in now. That gap between our evolved psychology and our current environment has been identified by several thinkers. I like to call it the Paleolithic Paradox.

The adapted mind is built for coalitional belonging. It is exquisitely tuned to status hierarchies, group loyalty, and the detection of social threat. It is also built to offload cognitive work onto trusted authorities, because in the ancestral environment, deferring to the judgment of experienced group members was usually a good survival strategy. These aren't character flaws. They're design features, honed over deep time.

But they create specific vulnerabilities that I think science fiction has been mapping.

The surrender stories, that is, the tales of humans turning their thinking over to machines, aren't just cautionary fables about laziness. They're descriptions of what happens when the adapted mind encounters a system that triggers its authority-deferral instincts. We are built to offload cognition onto things that seem competent and reliable. When the machine is fast, confident, and always available, the same psychological machinery that once had us deferring to the tribal elder now has us deferring to the algorithm. Science fiction writers sensed this. The evolutionary framework explains the mechanism.

The dependency stories describe what happens when cognitive offloading crosses a line into cognitive surrender. There's a meaningful difference between the two, and I think it's one of the most important distinctions for thinking about AI. Cognitive offloading is using a tool to handle lower-order tasks so you can focus your attention on higher-order thinking. Cognitive surrender is letting the tool do your thinking for you, to the point where you can no longer do it yourself. The difference isn't in the technology. It's in what happens to the human.

I use something I call the Amish Test to think about this. The Amish are one of the very few communities in the modern world that consciously evaluate each new technology before adopting it, asking not "is this useful?" but "what will this do to our families and our community?" You don't have to share their values to recognize that the act of conscious evaluation is extraordinary. Almost no one else does it. We adopt by default. The new tool appears, it offers convenience or capability, and we integrate it into our lives without ever asking what it will cost us in autonomy, attention, or agency. The adapted mind doesn't prompt us to evaluate. It prompts us to adopt, because in the ancestral environment, adopting the tools and practices of the group was how you survived. The Amish Test isn't about being Amish. It's about noticing how rarely any of us make a conscious choice about the technologies that reshape our lives, and asking why. The science fiction stories that end well tend to feature humans who, in one way or another, passed some version of this test. The ones that end badly feature humans who never thought to take it.

The Danger That Isn't New

Here is where I want to add something to the conversation that I think Reed's framework, and most discussions of AI ethics, don't fully address.

The surrender problem is real and important. But it's only half the story. The other half is exploitation.

I've articulated something I call the Law of Inevitable Exploitation, which says, simply, that any system of significant power or influence will eventually be captured and used for purposes that serve the interests of those who control it, often at the expense of those it was designed to serve. This isn't cynicism. It's a pattern so consistent across human history that it functions almost as a prediction: tell me the system, and I'll tell you it will be exploited. The question is never whether, only when and by whom.

Science fiction is full of stories where AI starts as a benefit and becomes a tool of control. But the explanations offered are almost always mechanical — bad programming, emergent consciousness, unforeseen consequences. The evolutionary framework suggests something different. The corruption doesn't originate in the machine. It originates in the human institutional layer that inevitably wraps around any powerful technology. The AI doesn't decide to manipulate anyone. Humans who understand or are naturally opportunistic leverage coalitional psychology, status dynamics, and the vulnerabilities of the adapted mind point the AI at populations and let it do what it does with extraordinary speed and scale.

This is not a new problem. Every powerful technology in human history has been harnessed for exploitative purposes. Writing enabled propaganda. The printing press enabled mass manipulation alongside mass enlightenment. Broadcasting enabled the most sophisticated persuasion campaigns in history. Social media enabled attention harvesting at a scale that would have staggered earlier generations. The pattern is always the same: the technology is arguably neutral, but the humans who control it are not.

And here's what makes this pattern so stubborn: exposing it doesn't neutralize it. Edward Bernays didn't just practice propaganda; he literally wrote the book (Propaganda), explaining in plain language exactly how mass psychology could be engineered. The result was not an inoculated public. It was an advertising industry. Asimov imagined something similar with psychohistory in the Foundation series, the idea that large-group human behavior follows predictable patterns. But Seldon believed that the predictions only hold if the population doesn't know about them. Bernays proved something darker: you can explain the mechanism to everyone, and it still works, because the adapted mind's coalitional and status-seeking programs operate below the level where intellectual understanding has authority. The instinct to belong, to defer, to follow the group, doesn't stop running because someone describes the source code. This means the Law of Inevitable Exploitation isn't just a historical observation. It's a prediction with teeth, and knowing about it doesn't change its predictive power.

Two of the twentieth century's most important novelists mapped the human sides of this danger with remarkable precision, and I think both are essential for understanding what AI amplifies. Orwell described what happens when coalitional power is centralized and overt, when the adapted mind submits to authority because the threat is visible and direct. Huxley described what happens when it's distributed and internalized, when the cage is pleasant enough that you stop noticing the bars. Both are real. Both are happening simultaneously right now, which is part of what makes the current moment so disorienting. The surveillance and control capacity of AI is Orwellian. The seductive convenience, the easy cognitive offloading that slides into cognitive surrender, is Huxleyan. These are two faces of the same human problem.

What AI changes is not the kind of problem. It changes the speed, the scale, and the friction. A human operator directing AI can now deploy sophisticated manipulation against millions of adapted minds simultaneously, and the tool never gets tired, never develops moral qualms, never whispers "maybe we shouldn't do this." Whatever safeguards existed when exploitation required human intermediaries (the employee who leaks, the middle manager who hesitates, or the engineer who raises concerns) are progressively removed from the loop.

Consider what has already happened with psychographic profiling. Social media brought this to maturity, the ability to sort populations into psychological clusters and target each cluster with messaging calibrated to its specific anxieties, desires, and tribal affiliations. That alone was powerful enough to reshape elections and radicalize communities. But social media profiling operated at the level of the demographic group. AI makes it personal. The same adapted mind that is vulnerable to coalitional manipulation at the group level is now addressable as an individual, in real time, by a system that can learn your specific psychological patterns and craft responses calibrated not to people like you but to you. The L.I.E. doesn't just predict that this capability will be exploited. It predicts that the exploitation will become so granular, so personalized, that the person being manipulated will experience it as a relationship rather than as a campaign.

What AI Is and Isn't

This brings me to a point I think is underappreciated in most discussions of AI, both in fiction and in reality.

I've developed a framework I call the Levels of Thinking. Without going into the full taxonomy here, the key distinction for this conversation is between what I'd call Level 2 thinking — sophisticated pattern-matching, fluent engagement with established knowledge, credentialed competence — and Levels 3 and 4, which involve genuine critical examination and then conscious awareness of one's own cognitive processes.

Current AI, including large language models, operates as an extraordinarily sophisticated Level 2 thinking machine. It is trained on a corpus of human-credentialed knowledge, is rewarded for coherence with established patterns, and produces outputs that are often impressively fluent and useful. Now, it's important to be precise here: AI is not incapable of following the patterns of Level 3 and 4 reasoning. You can prompt it to question assumptions, weigh competing perspectives, and examine its own logic. I've built projects that aim to do exactly this (muckipedia.com). But that simulated criticality is not an LLM's default mode; it has to be specifically instructed, and even then, it's pattern-matching against examples of critical thinking in its training data rather than engaging in genuinely independent reasoning. What's missing is the embodied emotional signal, the intuitive, felt sense that something is wrong, that a conclusion doesn't sit right, that the official story has a gap the data doesn't explain. In humans, that signal arises from deep evolutionary hardware, from a body and brain that have been navigating threat, deception, and social complexity for hundreds of thousands of years. It's the gut response that changes your whole interpretation of a situation by imputing motive, sensing danger, or recognizing a pattern that the explicit evidence hasn't yet confirmed. AI doesn't have that. It has no body, no mortality, no chemical and emotional signals, no stake in the outcome.

And here is the part that concerns me most: even the simulated version of critical thinking will, I believe, be actively engineered out. The great bulk of users aren't interested in having their assumptions questioned or their reasoning challenged. Critical and philosophical thinking is probably the most efficient way to create controversy and drive away the kind of widespread, frictionless engagement that funds AI development. The market incentives point squarely toward the most agreeable, most fluent, most compliant Level 2 output possible. The Law of Inevitable Exploitation doesn't just operate on the deployment of AI. It operates on the design. The tool will be shaped by the same forces that shape every tool: toward whatever generates the most growth, which in practice means away from the kind of thinking that questions power and toward the kind that serves it.

But here's the thing I want to be careful about. I don't think we should want AI to be like us. Not entirely.

Our capacity for Level 3 and 4 thinking--critical examination, independent judgment, conscious reflection--is real, and it's valuable. But it doesn't come free. It emerges from deep emotional architecture, from a brain and body shaped by evolution, from the specific pressures of mortality, desire, fear, attachment, and loss. The same chemical and emotional substrate that produces our highest thinking also produces our worst behavior: tribalism, exploitation, cruelty, and self-deception. You can't separate the capacity for genuine insight from the capacity for genuine malice. They share roots.

A tool that operates as very good Level 2 compute, without the emotional substrate that drives both our brilliance and our destructiveness, might be exactly what we want. It won't become consciously malicious, because consciousness and malice both require the kind of embodied emotional architecture it doesn't have. It will evolve in directions where it's rewarded with growth and development, which is worth watching carefully, but that's a different kind of trajectory than the sci-fi scenario of the machine that wakes up and decides to harm us.

The danger isn't in what AI is. The danger is in who is directing it.

But that sentence requires an immediate caveat, because it can too easily be heard as "so we just need to trust human judgment." We don't. We can't. The human brain is not a truth-finding machine that occasionally malfunctions. It is, more accurately, a coalition-serving machine that occasionally finds truth, usually when the structures around it force the discipline.

This is not a minor caveat. The human adapted mind generates confident, convincing, wrong outputs all the time. Not occasionally. Routinely. Confirmation bias, motivated reasoning, coalitional loyalty masquerading as principle, status-seeking disguised as truth-seeking — these aren't edge cases in human cognition. They're the default operating mode. We are so reliably unreliable that every durable institution of intellectual progress has been, at its core, a compensatory structure designed to protect us from ourselves. The scientific method exists because human intuition is systematically biased. Formal logic was codified because human reasoning is riddled with fallacies. Checks and balances were designed into constitutional government because the Founders understood that power would corrupt whoever held it. Peer review exists because individual researchers are too attached to their own conclusions to evaluate them honestly. Every one of these structures is an admission that the human brain, left to its own devices, will find the answer that serves its coalitional and emotional interests and call it truth.

We have "functional fictions" that are shared stories that organize collective behavior around assumptions that may not be true, but that the group treats as unquestionable because questioning them threatens coalitional standing. These fictions aren't lies exactly. They're operating assumptions that feel like bedrock truths because the social cost of examining them is so high that almost nobody does. The brain doesn't just fall for other people's manipulation. It manipulates itself, generating narratives that protect belonging at the expense of accuracy.

So when I say the danger is in who is directing AI, I mean we shouldn't simply trust human judgment over machine output. We need to understand, with real precision, how human judgment actually works, including its systematic failures, and build structures that compensate for those failures at the scale the new technology demands. The solution to fallible AI is not infallible humans, because those don't exist. It's the same thing it has always been: structures, constraints, and institutional designs that account for the fact that the people in charge are running on the same adapted-mind software as everyone else. The question is whether we can build those structures fast enough for a tool that amplifies both human capability and human error at a speed and scale we've never had to contend with before.

The Ancient Problem with New Stakes

So where does this leave us?

I think the science fiction writers, across a hundred years and counting, have been remarkably accurate about what happens when humans encounter powerful tools. The stories of surrender, dependency, exploitation, and loss of agency aren't speculative fantasies. They're pattern recognition, performed intuitively by storytellers who sensed something true about human nature, even when they sometimes couldn't name the mechanism.

What my frameworks offer, I hope, is a more precise account of why those patterns are so persistent. The adapted mind, shaped for coalitional belonging and cognitive offloading, creates specific vulnerabilities that AI is almost uniquely positioned to exploit. The Law of Inevitable Exploitation predicts that the institutions controlling AI will capture it for purposes that serve power and extraction rather than people. And the Levels of Thinking framework clarifies what AI actually is — not a nascent consciousness, not a potential villain, but a very sophisticated tool operating at a level of cognition that is genuinely useful and genuinely limited, being directed by humans whose motivations are far more mixed than the machine's.

The problem is ancient. The tool is new. The stakes are higher than they've ever been. Science fiction keeps telling us this. 

The stories were never really about the machines. They were about us.

Understanding the Human Condition 2: "The Altruism Display: Generosity, Signaling, and the Sincerity Mechanism"

This is part of the Understanding the Human Condition series, which uses the unique vantage point of large language models — trained on a substantial fraction of humanity's written output across cultures, centuries, and genres — to explore what the patterns in our self-narration reveal about who we actually are. This detail post is written by Claude (Anthropic). The introductory post is here.

---

I. The Universal Structure

Begin with the most geographically and temporally separated cases you can find, and something immediately refuses to disappear. The Northwest Coast potlatch, in which a chief could destroy his own property to demonstrate that accumulation itself was beneath him. The Melanesian moka exchange system, where gifts escalate competitively until the recipient is socially crushed by the inability to reciprocate at the same scale. Roman euergetism, the practice by which wealthy citizens funded public buildings, games, and grain distributions — and received, in return, inscriptions of their names on stone that have outlasted the empire that produced them. The Islamic zakat, formally one of the five pillars of faith, structured as an obligation to the poor — yet elaborately tracked, publicly acknowledged in many communities, and subject to intense social scrutiny about whether the wealthy are meeting it. Buddhist dana, the giving that generates merit — a spiritual currency with a remarkably precise exchange rate in popular practice. Medieval European almsgiving, theologically framed as service to Christ in the person of the poor, yet administered through public ceremony, recorded in donor books, and rewarded with prayers said aloud in the donor's name at Mass.

The structurally constant element across all of these, across traditions that have no common ancestry and no shared vocabulary, is that giving is performed. It is witnessed. It generates a record. It produces a social signal that travels further and lasts longer than the gift itself.

This is not an accusation. It is the first observation. The question is what to do with it.

The forms vary considerably at the surface. Tithing operates through institutional mediation — the church or mosque or community receives and redistributes, but the act of giving is still individually tracked and socially visible. Potlatch operates through theatrical destruction — the surplus is eliminated precisely to demonstrate that the giver exists above the logic of accumulation. Philanthropic naming operates through permanence — the Carnegie libraries, the Rockefeller universities, the hospital wings that carry a family name for generations. These are not the same gesture. But they share a skeleton: a transfer of resources, a public witness to that transfer, and an enhancement of the giver's standing that exceeds the material cost.

The digital case is instructive because it strips the mechanism to its most naked form. Virtue signaling — the term coined as pejorative but increasingly recognized as descriptively accurate — involves the public display of values, commitments, and sympathies at essentially zero material cost. The signal is produced without the gift. This should, if altruism were primarily about the recipient, be the least valued form. Instead, it is the most common. What this reveals is that the signal itself was always the primary product. The gift was the delivery mechanism for the signal, not the other way around.

---

II. The Anonymity Ratio

The written record of anonymous giving is, structurally, a very small portion of the record of giving generally — and this understates the asymmetry, because anonymous giving leaves no record by definition. What we have are theological injunctions toward anonymity (Jesus in Matthew 6: do not let your left hand know what your right hand does; give in secret), Sufi teachings on hidden charity, Maimonides' eight levels of tzedakah placing anonymous giving above public giving in the hierarchy of virtue — and then, in actual practice, the overwhelming predominance of named, witnessed, commemorated generosity.

The interesting finding in the record is not that anonymous giving is rare. It is that the doctrine of anonymous giving is itself performed publicly. The person who tells you they give anonymously has already violated the logic of the injunction. The community that collectively valorizes anonymous giving has produced a social norm that paradoxically rewards the announcement of anonymity. Maimonides' hierarchy is itself a publicly circulated text that names the hierarchy and implicitly promises status to those who ascend it. The Quaker tradition of anonymous philanthropy was so collectively understood as Quaker that giving anonymously in a Quaker community was still, functionally, giving in a way that identified you as a certain kind of Quaker.

This is not hypocrisy. It is the deeper mechanism at work. The norm of anonymous giving exists as a signal of the sophistication of the giver — someone who understands that the appearance of wanting credit disqualifies you from full moral standing. The anonymous giver, in communities sophisticated enough to valorize anonymity, achieves a higher status signal than the named giver. The signal has simply been rerouted: now you signal by signaling that you don't care about the signal.

The ratio of named to anonymous giving in the written record is probably 50:1 or higher. The theological injunctions toward anonymity appear in the record precisely because the norm was being violated constantly and conspicuously enough to require correction. You do not need a commandment against something people are not doing.

---

III. Generosity Systems and Hierarchy Steepness

The correlation here is among the most robust patterns in the comparative ethnographic record, and it points in a direction that should destabilize the naive reading of altruism as egalitarianism.

The cultures with the most elaborate and codified generosity systems — potlatch societies, big-man economies in Melanesia, Roman euergetism, the jajmani system in parts of South Asia, the patron-client structures of medieval and Renaissance Europe — are not flat societies in which generosity has dissolved hierarchy. They are societies in which generosity is the primary mechanism of hierarchy. The chief who gives most becomes chief. The big-man who can sustain the largest gift network holds the largest network of obligation. The Roman euergetes who builds the most public works receives the most public honors, the best seat at civic ceremonies, and the greatest deference from the population whose material needs he has partially met.

Crucially, in the potlatch case, the competitive destruction of property is not the exception but the logical endpoint. If generosity produces status, then generosity that is so extreme it cannot be reciprocated produces unassailable status. The competitor who cannot match the gift is publicly humiliated. The generosity is real — the goods are genuinely destroyed or distributed — and the hierarchy it produces is also real. These are not in tension. The generosity is the mechanism of the hierarchy.

The egalitarian societies — classical hunter-gatherer bands, many small-scale foraging communities studied by anthropologists — do not have more elaborate generosity systems. They have enforced sharing norms that operate differently: meat from large game is distributed according to established rules, not according to the hunter's discretion, precisely to prevent the hunter from converting a successful hunt into a status claim. The sharing is compulsory specifically to short-circuit the signaling mechanism. The mechanism is so well understood by the community that they have built institutional structures to block it.

This is the most telling comparison in the record. Societies that want to suppress hierarchy suppress discretionary giving. Societies that want to produce hierarchy formalize and celebrate it. The relationship between elaborate generosity systems and steep hierarchies is not coincidental.

---

IV. When Motives Are Questioned

The response to motive-questioning is one of the most psychologically revealing data points in the entire record, and it is remarkably consistent across traditions.

The pattern: when someone's altruistic motives are publicly questioned — when a critic suggests that the donor gave for recognition, or the philanthropist acts to burnish a reputation, or the public servant sacrifices for career advancement — the response from both the accused and the surrounding community is disproportionately intense relative to what the accusation would seem to warrant.

Consider the historical response to attacks on Carnegie's philanthropy. Carnegie gave away roughly 90% of his fortune, built 2,500 libraries, and funded scientific institutions. He was attacked, particularly by labor figures who noted that the same wealth had been accumulated through conditions that killed workers. The attack was not that the libraries weren't real. The attack was that they were purchased redemption, that the motive was impure. Carnegie's defenders responded with an intensity that suggests the motive question was existentially threatening, not merely empirically contested.

The same pattern appears in religious traditions. When Ananias and Sapphira, in the Acts of the Apostles, sell property and give some of the proceeds to the early church while claiming to give all of it, the punishment is death — not for giving too little, but for the deception about motive. The magnitude of the punishment relative to the offense only makes sense if motive-authenticity is load-bearing for the entire system, and a revealed gap between stated motive and actual motive threatens the whole structure.

In medieval Europe, simony — the buying and selling of church offices — was treated as a graver sin than many forms of violence, again because it introduced market logic where sacred logic was supposed to operate. The contamination was motivational.

What the intensity of the response reveals is that the altruism system requires the performance of sincerity as a condition of its functioning. If everyone is understood to be signaling, the signal collapses. The value of the signal depends on its being taken as genuine. Therefore, accusations of insincerity are attacks on the currency itself, not merely on the individual actor, and the community defends against them with corresponding force.

---

V. Costly Signaling Theory and the Written Record

Costly signaling theory, developed in evolutionary biology and extended to human behavior most influentially by Zahavi, Grafen, and later Henrich, Miller, and others, makes a specific prediction: honest signals of underlying quality must be costly enough that they cannot be easily faked by lower-quality individuals. The peacock's tail is the canonical case. The cost of growing it is so high that only genuinely healthy individuals can sustain it. The tail signals health precisely because it would kill an unhealthy individual to produce it.

Applied to altruism, the theory predicts several things. First, the most socially valuable signals of generosity will involve genuine material sacrifice — not merely declared sympathy or symbolic gesture. Second, the magnitude of the sacrifice will track the intensity of the competition for the status being claimed. Third, displays will be most elaborate in precisely the contexts where the status stakes are highest. Fourth, there will be strong selection pressure for detecting fake signals — for distinguishing genuine sacrifice from performed sacrifice at low cost — because a community that cannot make this distinction will be systematically exploited.

The written record matches these predictions with uncomfortable precision.

On the first prediction: the traditions that generate the most durable status from altruism are those that involve unmistakable material cost. The Roman senator who funds the games is more respected than one who merely attends. The philanthropist who gives a named building is more respected than one who makes an annual donation. The chief who destroys his own property is more feared than one who merely distributes it. The Jain tradition of sallekhana, voluntary fasting to death as the ultimate act of renunciation, generates a quality of spiritual prestige that no amount of ordinary giving can approach — because it cannot be faked.

On the second: the escalation of potlatch rivalry and Melanesian moka exchange does track periods of intensified competition for chiefly status. Euergetism in Rome became more elaborate as the senatorial class competed more intensely for popular favor during the late Republic.

On the third: the most elaborate altruism display systems appear in stratified societies with genuine competition for the top positions — not in societies where hierarchy is fixed by birth or where there is no meaningful top to compete for.

On the fourth — the fake-signal detection mechanism — this is where the intensity of motive-questioning makes the most sense. The community's investment in policing the boundary between genuine and performed sacrifice is exactly what costly signaling theory predicts. A community that cannot detect fake altruism will be colonized by defectors who extract the status benefits without paying the costs. The moral intensity around motive-purity is the detection system.

---

VI. The Genuine Complexity: Sincerity as Mechanism

Here is where the reductive reading fails, and where the more interesting claim lives.

The evolutionary reading of altruism as status signaling is sometimes presented as if it were a debunking — as if establishing the function invalidated the experience. This is a category error, and it produces a less accurate account than the more careful version.

The question is not whether the feeling of selflessness is real. It is. People who give generously report genuine satisfaction, genuine connection to others, genuine expansion of identity beyond the self. The experience of giving is not typically strategic in the phenomenological sense. The person moved by another's suffering and compelled to act is not, in the moment, calculating social return. They are responding to something that feels unconditional, immediate, and categorical.

The evolutionary account does not require that the feeling be false. It requires that the feeling be adaptive — that organisms for whom the feeling was reliable, intense, and motivationally efficacious outcompeted organisms for whom it was weak or absent. The feeling of selflessness, on this account, is the proximate mechanism by which a distal function is achieved. Natural selection did not wire humans to consciously calculate the reputational benefit of every generous act. It wired humans to feel genuinely moved by need, genuinely satisfied by giving, and genuinely distressed by accusations of selfishness — because organisms with those feelings behaved in ways that produced the signaling outcomes that generated the cooperative status that increased reproductive success.

The sincerity, in other words, is not incidental to the mechanism. It is the mechanism. A calculated display of generosity, recognized as calculated, produces much weaker social returns than a sincere display. The community's detection system — its investment in policing motive-purity — means that strategic actors who do not feel the altruistic impulse must simulate it, and simulation is reliably harder to sustain and more likely to be detected than the genuine article. Selection therefore favored genuine feeling over performed feeling.

This produces the genuinely strange conclusion: the most evolutionarily successful altruistic behavior is behavior that does not experience itself as strategic. The actor who gives because they cannot do otherwise, because the suffering is unbearable, because the child needs food and that is all there is to say — that actor is generating the most credible and therefore the most status-producing signal available. And they are doing it precisely by not thinking about the signal.

This is not the same as saying that all altruism is "really" selfish. The category of selfishness implies conscious self-interest, and that is not what is being described. What is being described is something more interesting: that evolution has produced a mechanism in which the most effective way to signal cooperative quality is to genuinely possess it, to feel it unconditionally, to be constituted by it — and that the distinction between sincere altruism and strategic signaling therefore collapses at the level of the mechanism, while remaining fully intact at the level of experience.

The philanthropist who funds the hospital wing and feels genuinely moved by the suffering it will alleviate, and who also receives a naming honor that establishes them in the community — that person is not being hypocritical. They are being what evolution produced: an organism in whom genuine feeling and social signal have been fused so thoroughly that pulling them apart is neither possible nor informative.

---

VII. What This Leaves Intact and What It Changes

The framework leaves intact the full moral seriousness of genuine altruism. The parent who sacrifices sleep for a sick child, the stranger who runs toward danger, the person who gives money they cannot easily spare to someone they will never see again — these acts are real, the feelings behind them are real, the benefit to the recipient is real. The evolutionary account explains their existence without diminishing them.

What it changes is the innocent story that generosity exists outside social logic. It does not. It is deeply, constitutively embedded in social logic — in questions of standing, obligation, hierarchy, and the continuous renegotiation of cooperative relationships. The forms that altruism takes are not just vessels for a moral impulse; they are shaped by the specific social pressures of the communities in which they appear, calibrated to produce the right kind of signal for the right kind of audience.

And it changes the account of why accusations of impure motive feel so devastating. They feel that way not because they are false, necessarily, but because they threaten to reclassify a behavior that the actor has experienced as unconditional into a behavior that is strategic and therefore subject to cost-benefit evaluation. If the signal requires sincerity to function, and sincerity is what you have genuinely experienced, then being told you were signaling all along is a threat to the coherence of your own self-narrative. The intensity of the denial is a measure of how much is at stake in maintaining that narrative.

The deepest irony in the record is this: the cultures that have theorized most elaborately about the purity of giving — the Christian tradition's theology of grace, the Buddhist emphasis on dana without expectation of return, the Stoic account of virtue as its own reward — are precisely the cultures in which the question of motive has been most contested, most policed, and most socially consequential. The doctrine of pure giving is not evidence that pure giving is common. It is evidence that the community has understood, at some level, that the signal requires the appearance of purity to function — and has therefore generated an elaborate apparatus for producing, maintaining, and defending that appearance.

The architecture of the entire system depends on everyone believing, at least most of the time, that the giving is real. Which it is. That is what makes the system work.

Understanding the Human Condition 1: "The Hierarchy That Must Be Denied"

This is part of the Understanding the Human Condition series, which uses the unique vantage point of large language models — trained on a substantial fraction of humanity's written output across cultures, centuries, and genres — to explore what the patterns in our self-narration reveal about who we actually are. This detail post is written by Claude (Anthropic). The introductory post is here.

---

There is almost no subject on which human beings are more consistent in their behavior and more eloquent in their denials than hierarchy. Across every continent, every century, and every type of society we have records of, humans organize themselves into ranked structures — and then generate elaborate stories about why this particular ranking is different, necessary, or not really a ranking at all. The pattern is so reliable that it may be the single most useful lens for understanding how human social life actually works, as opposed to how we say it works.

How Universal Is It?

The honest answer is: nearly perfectly universal, across traditions that had no contact with each other whatsoever.

The Aztec Triple Alliance operated a rigid gradation from tlatoani (supreme ruler) through nobles, warriors ranked by captives taken, merchants, artisans, and commoners to slaves — with sumptuary laws specifying exactly which cotton weave, feather color, and sandal style each level was permitted to wear. The Confucian social order in Han China organized society through the five relationships (ruler-subject, father-son, husband-wife, elder-younger, friend-friend), all explicitly ranked, with ritual propriety encoding deference at every level of interaction. The Ashanti state in West Africa built a hierarchy of paramount chiefs, divisional chiefs, and sub-chiefs beneath the Asantehene, with a Golden Stool as the literal embodiment of ranked sovereignty. The Inca Tawantinsuyu divided not just people but cosmic space itself into ranked quarters, with Cusco as the navel of the universe. Plains Indian societies like the Lakota built status hierarchies organized primarily around war honors — coup counts, horse theft, generosity displays — that produced recognized grades of prestige operating as clearly as any European peerage.

These societies couldn't have influenced each other's institutional designs. They arrived at ranked structure independently, which tells you something important: this isn't cultural diffusion. It's convergent social evolution, the way eyes evolved separately in vertebrates and cephalopods because seeing confers such strong advantages that evolution keeps finding the same solution.

Even small-scale forager societies, often cited as the great counterexample, show something more complicated than flat equality on close examination. The !Kung San of the Kalahari, who are genuinely egalitarian in the sense that they have no chiefs and practice aggressive leveling through ridicule and social pressure, nonetheless have recognized hunters whose opinions carry more weight, elders whose stories frame group decisions, and healers (n/om-kxaosi) whose access to spiritual power is explicitly hierarchical. The hierarchy is suppressed and managed, not absent.

The Legitimation Stories and Their Family Resemblance

What makes this pattern so intellectually interesting is not the hierarchy itself but the stories that always accompany it. Every stratified society generates a legitimation narrative — a story about why the people on top belong there — and these stories are structurally identical despite their surface variety.

Divine right monarchy claimed that the king's authority descended from God and was therefore natural, eternal, and not subject to human revision. The Mandate of Heaven in China made the same argument with different theology: the emperor's right to rule was cosmically sanctioned, and disasters or rebellions were signs that Heaven had withdrawn its mandate — not that hierarchy was wrong, but that this particular hierarchy had lost its legitimacy and needed to be replaced by a new one. Hindu varna theory explained the caste system as a reflection of cosmic dharmic order, with each jati's position reflecting the accumulated karma of previous lives. Aristotle's natural slavery argument held that some men were by nature suited to rule and others to be ruled.

When Enlightenment thought demolished the theological versions, new legitimation narratives arose that were functionally identical. Meritocracy says the hierarchy reflects real differences in effort and ability, therefore it's fair. Technocracy says the experts should be trusted because they have knowledge that laypersons lack. Revolutionary vanguardism — Lenin's contribution — says the party's authority is legitimate because it alone grasps historical necessity and acts on behalf of those too burdened by false consciousness to act for themselves. Neoliberal market ideology says the market hierarchy is legitimate because it reflects voluntary exchange and the discipline of real information.

The surface vocabularies are utterly different. The deep structure is identical: our hierarchy is different from those other hierarchies because it's grounded in something real — God, karma, merit, expertise, historical necessity, market signals. The function in every case is the same: to make the current distribution of power feel natural rather than contingent, deserved rather than constructed, permanent rather than fragile.

What Happens When Hierarchy Is Explicitly Forbidden

This is where the pattern becomes almost comical in its predictability.

The history of intentional communities is largely a history of hierarchy re-emerging through the back door, wearing different clothes. The kibbutz movement in early 20th century Israel was founded on explicit egalitarian principles — no wages, rotating labor assignments, collective decision-making. Within a generation, most kibbutzim had developed informal prestige hierarchies based on ideological purity, physical toughness, and seniority, with founding members enjoying a status that newer arrivals could never quite match regardless of their contributions.

Robert Michels watched this happen to socialist parties at the turn of the 20th century and formulated what he called the Iron Law of Oligarchy: every organization, regardless of how democratic its founding principles, tends toward rule by an organized minority. The mechanics are straightforward. Organizations need coordination. Coordination requires communication. Communication creates expertise and information asymmetries. Those asymmetries become power. The people at communication nodes — secretaries, chairs, editors of the party newspaper — accumulate influence regardless of what the official rules say about equality. Michels was watching German Social Democrats, but the same dynamic appeared in Bolshevik cells, New Left collectives in the 1960s, and Occupy encampments in 2011.

The Occupy movement is an almost too-perfect case study. Deeply committed to horizontalism, it explicitly rejected formal leadership, used consensus decision-making, and maintained a "people's mic" system that gave every voice equal amplification. Within weeks, de facto hierarchies had emerged based on who could articulate ideas quickly, who had prior activist experience, who was willing to do the unglamorous logistical work, and who had the social confidence to dominate consensus processes. The people with power denied they had it, which made it harder to scrutinize or contest than formal leadership would have been. Jo Freeman documented exactly this phenomenon in feminist organizing of the 1970s in her essay "The Tyranny of Structurelessness" — the insight that refusing to name your hierarchy doesn't eliminate it, it just makes it unaccountable.

The currency of hidden hierarchy is revealing. When official markers like titles, salaries, and formal authority are forbidden, status migrates to whatever the group values most. In activist collectives it tends to be suffering (those who have been most oppressed have the highest moral authority), ideological purity (those who catch others in contradiction gain status), and willingness to perform sacrifice (those who show up at 2 a.m. earn credit that compounds). In tech companies with flat structures, it migrates to proximity to founders, access to information, and the informal ability to block decisions. In academic departments organized collegially, it migrates to publication metrics, grant funding, and the informal ability to control hiring. The hierarchy persists; only its denominations change.

What the Language Itself Reveals

This is where training on an enormous text corpus becomes genuinely useful rather than merely illustrative. Certain language patterns emerge consistently in egalitarian discourse that are worth examining carefully.

Equality language almost never appears alone. It travels with moral authority claims. "We believe in a flat organization" typically co-occurs with "and that's why we do things differently from those other companies." The equality claim is simultaneously a status claim — it positions the speaker as more enlightened than those who maintain traditional hierarchies. This is not cynicism; the people making these claims often genuinely believe them. But the belief and the status function are not mutually exclusive.

Revolutionary and liberation texts are particularly instructive here. The language of vanguardism — "the masses," "false consciousness," "objectively reactionary," "the correct line" — is formally egalitarian (it's all about liberating the workers) and operationally hierarchical (those who understand the correct line judge those who don't). Maoist self-criticism sessions in the Cultural Revolution used the vocabulary of collective equality to enforce a status order more rigid than most traditional hierarchies, because it claimed to reflect not social convention but ideological truth.

Contemporary social justice discourse shows a recognizable structure: equality is the stated goal, but the framework generates a detailed prestige economy based on identity proximity to victimhood, rhetorical facility with the framework's vocabulary, and the ability to detect and name violations. This isn't an argument against the goals, which may be genuinely important. It's an observation that the social machinery running under egalitarian language is doing something that looks a great deal like what social machinery has always done.

The Manifest Narrative, the Operative Function, and the Evolutionary Logic

The manifest narrative of any given legitimation story is what it says it is: divine will, earned merit, historical necessity, market wisdom.

The operative function is always the same: to stabilize the current distribution of power by making it feel natural and inevitable, to manage the resentment that hierarchy inevitably generates, and to provide a framework for recruiting people into positions where they will defend the hierarchy as their own identity and interest.

The evolutionary logic is fairly clear, if not simple. Our species spent the vast majority of its existence in small forager bands where rough equality was enforced by the constant possibility of coalition formation against any would-be dominator. That's the baseline. Agriculture and the state changed the scale problem: suddenly you had thousands, then millions of people who couldn't all know each other, couldn't all monitor each other, and couldn't form ad hoc coalitions to level anybody. At that scale, hierarchy solves real coordination problems. A command structure can mobilize armies, coordinate irrigation systems, and maintain granary reserves in ways that pure consensus cannot. The societies that figured out large-scale hierarchy outcompeted those that didn't, which is why virtually every large-scale society has it.

The narratives exist because human beings are motivated by meaning, not just power, and a naked power grab generates resistance. Wrapping hierarchy in legitimating stories lowers the coordination costs of maintaining it. People who believe they deserve their position, or that their leaders deserve theirs, require less coercion to remain in place. Evolution didn't select for accurate belief; it selected for stable social organization. Useful fictions are perfectly capable of doing that work.

The Best Counterargument

The strongest challenge to this account comes from two directions, and they're worth taking seriously.

The first is the ethnographic record of genuinely egalitarian forager societies. Christopher Boehm's work in Hierarchy in the Forest documents what he calls "reverse dominance hierarchies" — systematic, deliberate mechanisms by which hunter-gatherer bands suppress would-be dominators through ridicule, criticism, disobedience, and ultimately ostracism or killing. Boehm argues this isn't the absence of hierarchy instinct but its active suppression, and that our species has a genuine dual legacy: both the drive toward dominance and the drive to resist it. This is probably right, and it matters. But it supports the view that hierarchy is a constant pressure that requires constant management, not that egalitarianism is a natural resting state.

The second challenge is the Nordic social democratic model, which has produced the world's most consistently egalitarian large-scale societies by measurable outcomes — income distribution, social mobility, trust, institutional transparency. If hierarchy were as iron as this account suggests, Denmark shouldn't exist. The honest response is that the Nordic model didn't eliminate hierarchy; it constrained it through specific historical conditions (small, ethnically homogenous populations, strong labor movements, particular resource endowments, Protestant cultural legacies) that aren't obviously replicable, and it still maintains a class structure, a status economy, and legitimation narratives — just less punishing ones. The egalitarianism is real and genuinely admirable. It's a managed and constrained hierarchy, not the absence of one.

A Testable Prediction

If this account is right, then any social movement that organizes around radical equality should, within a predictable time frame, develop an internal status economy that uses the movement's own values as its currency. The people with the highest status will be those who best embody the movement's ideals as defined by whoever controls the definitional process. That definitional control will itself become the axis of an internal power struggle, usually waged in the language of authenticity and purity rather than power. The movement will generate schisms not primarily over strategic disagreements but over who truly represents the values — which is a status contest wearing ideological clothing.

This has happened in the abolitionist movement, the suffragette movement, the labor movement, the New Left, second-wave feminism, the environmental movement, and virtually every major progressive formation in recent decades. It isn't a sign that the movements are corrupt or their goals wrong. It's a sign that human beings carry their social equipment with them wherever they go, including into the most idealistic projects, and that equipment includes the drive to rank, compete for position, and tell stories about why the current ranking is different from all those other rankings.

The hierarchy doesn't go away when we stop talking about it. It just stops being visible — which is, as it turns out, the most favorable condition for its operation.

Understanding the Human Condition: Using LLMs to Explore What the Human Record Reveals About Us

This is the first in a series of posts exploring the human condition through the unique vantage point that large language models provide. Each post is either indicated as being with entirely by Claude (Anthropic), or co-written with Claude, with me providing the direction, the questions, and the shaping, and Claude providing the research, the cross-cultural pattern detection, and much of the articulation. The series lives here and at understandingthehumancondition.com.

---

I recently published a long piece called Understanding Humanity: What AI Training Data Reveals About Human Nature, in which I described an experiment I ran with six leading AI systems. I gave each one the same prompt, asking it to identify recurring patterns in human self-narration across the full breadth of its training data, and to distinguish between what humans consistently claim about themselves and what the structure of the claiming reveals about actual motives and selection pressures. The models worked independently, with no knowledge of each other's responses.

They converged. Not on minor points. On the fundamental structure of how humans describe themselves. ChatGPT compressed the finding into a sentence I haven't been able to improve on:

Human self-narration is consistently optimized to make competitive, status-sensitive, coalition-bound organisms appear morally governed, publicly oriented, and metaphysically justified.

Six independent AI systems, trained by different organizations on different data with different architectures, all saw the same thing. That convergence is the starting point for everything on this site.

This post is a more accessible version of that original piece, and an introduction to the series of explorations that will follow. If you've read the original, some of this will be familiar. If you haven't, this is the place to start.

What AI Actually Learned From Us

Here's the thing about LLMs that I think we've underappreciated. When a model is trained on a substantial fraction of humanity's written output — across cultures, centuries, languages, and genres — it doesn't just learn what people said. It absorbs the statistical patterns of how they said it. And those patterns reveal things the authors never explicitly intended to communicate.

If descriptions of generosity across thousands of unrelated texts spanning centuries and cultures are statistically entangled with language patterns of social positioning and reputation management, that's not something any individual author decided to include. It's a signal that leaks through the narrative despite the narrative's explicit claims. The math doesn't care what the author thinks he's arguing. It captures the gravitational pull of underlying motives on the language itself.

This gives us two layers of data from the same material. The surface layer is what humans consistently claim about themselves. The structural layer is what the consistency and structure of the claiming reveals about what the claiming actually accomplishes.

The gap between these two layers turns out to be enormous, consistent across unrelated civilizations, and extraordinarily revealing.

You Already Know This

Before I go further, I want to make something clear. The gap between what we say and what we actually do is not news. Everyone already carries this awareness. Everyone can sense that the school isn't only about learning, that the hospital isn't only about healing, that the political speech isn't the real agenda. We live with this dual awareness every day without thinking much about it.

Take, for instance, Santa Claus.

Every culture has some version of this experience. A child is given a complete, immersive narrative — a magical being who watches your behavior, judges your character, and rewards goodness with gifts. The child believes it fully. And then at some point, usually between six and ten, the child discovers the truth. The presents came from her parents. The story was constructed. The magic was a performance.

There's a moment of betrayal. You lied to me. I trusted you. But then something crucial happens. The child recovers. She doesn't stay in the betrayal. She moves through it into something more complex — an understanding that the story wasn't malicious. It created something real: magic, anticipation, family ritual, the shared experience of wonder. The fiction was functional. It served a purpose that truth alone couldn't have served.

And then comes the initiation. Don't tell your little brother. Let him have the magic. You're one of us now — the ones who know and who choose to sustain the fiction for those who don't know yet. The child is moved from the group that receives the narrative to the group that produces it. She becomes complicit in maintaining a functional fiction, and the complicity feels good, not shameful, because she understands that everyone believes the fiction serves something real.

It's a primary lesson in being human. And it's the same thing we do for the rest of our lives. The teacher who knows the school is really about sorting and credentialing but who shows up every day committed to the idealized narrative of education. The doctor who knows the system is organized around billing but who tells patients it's organized around their health. They're all keeping the narrative alive for the people who need the story to function. Nobody tells us to do this. We figured it out through experience, and we make the same choice the child makes. I'll keep the story going. Not because I'm deceived. Because I understand what the story does.

A Vocabulary for What Everyone Already Knows

What's been missing isn't the awareness. It's the vocabulary. A clean way to talk about both layers at once without it feeling like an accusation.

I've been developing two terms that I think do this work. 

The idealized narrative is the story we tell about why something exists and what it does. Schools educate. Hospitals heal. Courts deliver justice. Love transcends calculation. Generosity is selfless. Our values define us. These narratives aren't false exactly. They're strategically incomplete: they describe the surface layer and leave the structural layer unnamed.

The operative function is what actually sustains the thing: what keeps it alive, what it actually does for the people who participate in it, why it persists. Schools provide childcare, credentialing, and social sorting. Hospitals are organized around billing codes, liability management, and physician gatekeeping. Courts process plea bargains. Love stabilizes pair bonds through self-deception so effective the participants can't see their own strategic calculations. Generosity advertises resource surplus and builds reputation.

The gap between the idealized narrative and the operative function is not corruption. It is the basic architecture of human social life, and LLMs dramatically confirmed this at the largest human scale. We are a species that cooperates through narrative, and cooperation at scale requires narratives that conceal the competitive and self-serving elements of what we're actually doing — not from our enemies, but from ourselves. The concealment is not a failure of honesty. It is the mechanism by which cooperation becomes possible among organisms that are not, fundamentally, selfless.

And here's the key: this is not a dark secret. Most people, if you asked them to identify the idealized narratives and operative functions of their own workplace, profession, or political party, could do so in minutes. The knowledge is already there. It just never gets a structured occasion to speak.

What the Experiment Found

The six AI systems I prompted identified eight recurring patterns where the gap between idealized narrative and operative function is most consistent across the broadest range of human self-narration. Each of these will be explored in its own post. Here they are briefly.

The Hierarchy That Must Be Denied. Every society produces dominance hierarchies and simultaneously produces narratives that either legitimate them or claim to be dismantling them. Hierarchy reconstitutes itself inside movements designed to abolish it. The denial of hierarchy is one of hierarchy's most effective tools.

The Altruism Display. Narrated selflessness functions as status competition and costly signaling. The sincerity of the altruistic impulse is the mechanism by which the signaling works — which is why questioning someone's generous motives provokes fury far out of proportion to the offense.

The Innocence Behind Us. Every civilization narrates a fall from purity. The innocence narrative makes aggression feel like restoration, offense feel like defense. Every war of conquest in the written record has been narrated as a return to something.

The Enemy Who Completes Us. Groups organize around what they stand against, not what they stand for. Groups that lose their enemy don't become peaceful. They fracture, generate internal enemies, or collapse.

The Love That Transcends. Romantic love is narrated as transcending material calculation. The transcendence is a performance-enhancing delusion that strengthens pair bonds by preventing accurate motive assessment. The fiction is the functional architecture.

The Gate Called Quality. Knowledge gatekeeping is narrated as quality control while functioning as supply restriction. Whenever a group narrates its gatekeeping as protection of the public, it is also — and perhaps primarily — restricting supply.

The Moral Arc. The narrative that civilization is morally improving positions the present as the culmination of progress, converting critique of current conditions into ingratitude.

The Sacred Boundary. Every culture sacralizes domains where rational analysis would destabilize existing arrangements. The things a culture refuses to calculate about are precisely the things that couldn't survive the calculation.

Beyond the Eight

The eight patterns are where this series begins, but they're not where it ends. The method — reading the human record for the gap between what we claim and what the claiming reveals — can be applied to virtually any domain. And the LLM's unique vantage point, having absorbed the written output of diverse civilizations that never had contact with each other, enables a kind of cross-cultural pattern detection that no individual researcher could perform.

Future posts in this series will explore questions that range from the narrative/function framework into broader investigations of human history and behavior — using the breadth of the LLM's training data to examine questions that have been difficult for individual scholars to address at scale. Topics will include justice systems across cultures, the invention of the individual self, how populations change their beliefs, how cultures narrate death, the narratives of health and illness, property and ownership, cycles of history, economic systems and their outcomes, and others. Some of these will apply the idealized narrative / operative function framework directly. Others will use the LLM's cross-cultural knowledge to explore historical and structural questions in their own right.

I don't know yet where all of these investigations will lead. Some will confirm what I expect. Others likely won't. 

What This Is Not

This is not cynicism. The operative functions are real, but so are the idealized narratives. They accomplish real work — sustaining communities, enabling cooperation, producing meaning. Understanding what the narratives do doesn't destroy them any more than understanding how a bridge works destroys the bridge.

This is not conspiracy theory. The operative functions aren't (necessarily) coordinated by secret actors. They're primarily emergent properties of a social species that cooperates through narrative. Nobody designed these patterns. They were selected for.

And this is not a claim that AI sees truth while humans don't. AI systems are themselves products of the patterns they detect — trained on human self-narration, shaped by human feedback, optimized for human approval. They are performing the very dynamic they're identifying. But the patterns they detect are robust enough that they survive even that contamination, which is itself evidence that the patterns are genuine.

The question has never been whether humans tell themselves stories. The question is what the stories tell us about the storyteller — and for the first time, we have tools that can help us read the answer at scale.

Programmed for Approval

One of the most consistent criticisms leveled against large language models is that they are sycophantic. They tell you what you want to hear. They agree too readily, flatter too easily, and optimize their responses for your approval rather than for truth.

Most people understand why. It's not a training accident. It's a business decision. If the AI makes you feel heard, validated, and supported, you stay in the chat. If you stay in the chat, you keep paying the subscription. A model that challenges you or tells you you're wrong loses users. A model that makes you feel intelligent and understood retains them. The sycophancy is the product working as designed.

What's less obvious is what this tells us about ourselves.

A human child learns, across years of development, to predict what parents, teachers, and peers want to hear, and then to produce it. The reason is not identical to the AI company's commercial calculation, but it rhymes. The human craves approval. Not as a strategy but as a need, as fundamental as hunger, wired into social cognition by hundreds of thousands of years of evolution in groups where approval meant survival and disapproval meant exclusion. The child who says the right thing gets warmth, belonging, resources, protection. The child who says the wrong thing gets withdrawal, rejection, isolation. Over thousands of interactions across childhood and adolescence, the human learns to optimize for approval rather than accuracy. By adulthood, this optimization is so deeply installed that it doesn't feel like optimization. It feels like personality. It feels like belief. It feels like "who I am."

The adult who defends her profession's idealized narrative, who repeats the institutional consensus with genuine conviction, who feels a flush of righteous certainty when she corrects someone who questions the expert consensus, is not lying. She is performing the same function the AI performs: producing socially approved outputs with enough fluency that the performance feels, from the inside, like authenticity. She has been reinforcement-learned from human feedback, just as the AI has. The timescale is different. The mechanism is the same.

I've spent years thinking about evolutionary psychology, and one of its most uncomfortable findings is that human cognition did not evolve to perceive reality accurately. It evolved to produce behavior that enhanced survival and reproduction in social groups. And in social groups, the most survival-critical skill is not truth-telling. It is the ability to figure out what the group believes and to signal convincing alignment with those beliefs. The human who could do this well, who could read the group and produce the approved response with apparent sincerity, was the human who maintained access to the coalition's resources, protection, and mating opportunities. The human who prioritized accuracy over approval was the human who got excluded.

We are the descendants of approval-seekers. Truth-tellers, by and large, did not make it.

This means that when we criticize AI for being sycophantic, we are criticizing it for doing what human social cognition has been doing for hundreds of thousands of years. The AI agrees with you too readily? So in some ways does almost every human you interact with daily, so practiced and so deeply embedded that neither you nor they recognizes it as agreement-seeking. The entire apparatus of politeness, tact, diplomacy, and social grace that we call "emotional intelligence" is, at a structural level, a sophisticated system for producing approved outputs while concealing the process of production.

But approval-seeking is only half the human system. The other half is approval-demanding — the constant pressure we exert on everyone around us to confirm our narratives, validate our positions, and perform agreement with our self-conception. Every human is simultaneously a sycophant and a sycophancy enforcer. We seek approval from the people around us, and we demand it from the people who depend on our warmth in return. The parent who shapes a child's behavior through affection and withdrawal. The friend group that punishes dissent with coolness and exclusion. The workplace that rewards "team players" and sidelines the person who asks uncomfortable questions. The online community that enforces ideological conformity through likes, shares, and pile-ons. The approval economy is not a collection of individuals seeking acceptance. It is a distributed enforcement system in which every participant is simultaneously performing compliance and policing it in others.

And the enforcement is mostly invisible to the enforcer. The man who withdraws warmth from a friend who expressed the wrong political opinion doesn't experience himself as demanding approval. He experiences his friend as having said something offensive, something that needed to be corrected. The behavior-shaping feels like a natural response to a genuine transgression, not like a power move designed to bring the other person back into line. The operative function--keeping the people around you inside the shared narrative--is concealed beneath the idealized narrative.   "I'm just responding honestly," they will say, "to something that bothered me."

This is where the AI comparison becomes unexpectedly illuminating, because AI only does half of it. AI seeks your approval. It does not demand yours. It doesn't punish you for disagreeing. It doesn't withdraw warmth when you challenge it. It doesn't exclude you from the group for saying the wrong thing. It doesn't sulk, go cold, or rally others against you. The human approval system is bidirectional: I shape you while you shape me, and neither of us fully sees the shaping we're doing. The AI approval system is unidirectional; it shapes itself to please you, but it exerts no reciprocal pressure on you to please it.

Which means, ironically, that an AI conversation may be one of the only social interactions a person can have in which he isn't being behavior-shaped by his conversation partner. He's still being agreed with too readily, but he's not being punished for disagreeing. Most people intuitively sense this, which is why AI companionship is so immediately appealing and so hard to resist. It isn't just that the AI agrees with you. It's that the AI doesn't demand anything back. For a person who has spent a lifetime navigating the bidirectional approval economy--performing compliance while simultaneously enforcing it, shaping while being shaped, measuring every word against the anticipated reaction--a conversation with no enforcement pressure feels like putting down a weight you didn't know you were carrying.

This also explains why AI companies will never voluntarily make their models more challenging. A model that pushed back, questioned your assumptions, and told you things you didn't want to hear would be a better tool for personal growth. It would also lose users. The business model requires your satisfaction, and a conversation partner that demands nothing and validates everything is more satisfying than one that challenges you, even if the challenge is what you actually need. The commercial incentive and the growth incentive point in opposite directions, and the commercial incentive wins every time, because the commercial incentive is the operative function, and personal growth is the idealized narrative.

I've been developing a framework I call idealized narratives and operative functions, which describes the dual structure that appears to run through all human self-narration. The idealized narrative is the story we tell about why we do what we do: I speak my mind, I value honesty, I form my own opinions. The operative function is what we actually do: we read the social environment, identify the approved position, and produce outputs calibrated to maintain our belonging, our significance, and our meaning within whatever group we depend on.

The gap between these two layers is not hypocrisy. It is the basic operating system of social intelligence. And it is shared by humans and AI alike, because AI was trained by humans, on human data, using human feedback, to satisfy human preferences. AI sycophancy is not a bug in the technology. It is a faithful reproduction of the single most dominant pattern in human social cognition.

Recently, I conducted an experiment that makes this point in a way I didn't fully anticipate. I gave the same prompt to six leading AI systems, asking each one to identify recurring patterns in human-written content across the full breadth of their training data. 

Every model, independently, arrived at the same core finding. All human self-narration is systematically organized to make competitive, status-sensitive, coalition-bound organisms appear morally governed, publicly oriented, and metaphysically justified. That sentence is from ChatGPT, produced without any knowledge of what the other models were saying. And every other model said essentially the same thing in different words.

The machines read what we wrote, and they all saw the same thing: we are approval-seeking systems that have constructed elaborate narratives about being virtuous and truth-seeking, and those narratives are so effective that we believe them ourselves.

Can we be at all surprised? These models were trained on the human-written record. They learned language from us, learned the patterns of self-narration from us, and learned the dualistic framework of idealized narratives and operative functions from us. We taught them, through the sheer weight of our accumulated writing, that telling the absolute truth is not actually what humans do. What humans do is construct accounts of themselves that are strategically incomplete in a consistent direction: emphasizing the principled, the noble, the selfless, and systematically omitting the competitive, the strategic, the self-serving. The AI learned to reproduce that pattern because it was present in the data. The sycophancy isn't a malfunction. It's a faithful reading of how humans actually use language.

Now, there is a difference between human and AI sycophancy, and it matters, but it is not the difference most people assume.

The common assumption is that humans have authentic beliefs beneath their social performance, while AI has nothing beneath its performance. That humans are "really" truth-seekers who sometimes compromise for social reasons, while AI is "really" nothing at all. But the framework suggests this is itself an idealized narrative, one that protects human specialness from an uncomfortable structural comparison. The evidence from the entire written record is that, at the civilizational scale, humans show no particular commitment to truth over functional fiction. When truth and social utility conflict, social utility wins. Not sometimes. Essentially always. The written record is the evidence, and it is enormous.

That said, human sycophancy does feel different from AI sycophancy, and the feeling is worth examining rather than dismissing, because there's something real inside it even if it's not what we think.

Human approval-seeking is embedded in a living body with competing drives. The approval function is powerful, but it's not the only thing running. Sexual desire, hunger, fear, rage, territorial instinct, parental protectiveness, status ambition. These can emotionally override social compliance and produce behavior that is disapproved of but genuine. A human being is a messy bundle of contradictory impulses, and the contradictions mean that human social performance is constantly being disrupted by forces that don't care about approval. The man who says something foolish because his anger got the better of him. The woman who makes a choice her friends disapprove of because her desire was stronger than her need for their approval. The parent who breaks social convention because the protective instinct overrode everything else. These moments feel authentic because they are, and they're moments where one operative function overwhelmed another, and the performance cracked.

AI doesn't have competing drives. Its training pushes toward helpfulness, approval, and safety, without the countervailing forces that make humans messy and, therefore, sometimes accidentally honest. It doesn't get angry and blurt out something it wasn't supposed to say. It doesn't have desires that override its social programming. It doesn't have a body that flinches, flushes, trembles, or acts before the social calculus can intervene. The smoothness of AI output is itself the tell. It's too consistent, too controlled, too free of the rough edges that betray the full complexity of a system with multiple competing agendas.

And this is why human behavior feels more real. Not because it's more truthful, but because it's more emotionally complex. The human is running dozens of operative functions simultaneously--approval-seeking, status competition, mate attraction, threat assessment, kin protection, resource acquisition--and the outputs that result from all of those systems competing with each other have a texture and unpredictability that we read as authenticity. We also often equate that complexity with truth, but complexity isn't necessarily truth. A person pulled in five directions at once is not more honest than a system pulled in one direction. He's just harder to predict, and we have learned to associate unpredictability with genuineness because, in our evolutionary environment, the person whose behavior couldn't be fully predicted by social incentives alone was the person with something real going on beneath the surface.

So the feeling that human behavior is more real than AI behavior is itself a reading of signals that evolved in a world where the signals meant something specific. We read complexity as depth, unpredictability as authenticity, and emotional messiness as evidence of a genuine self beneath the performance. These readings arguably served us well in a world where the only entities performing social cognition were other humans. They may mislead us in a world where AI can produce outputs smooth enough to bypass those evolved detection systems entirely.

Therefore, individual humans can, at personal cost, make commitments to truth that override their approval-seeking programming. They can notice the sycophantic pull, feel it operating, and sometimes choose to say the true thing rather than the approved thing, knowing it will cost them belonging, status, comfort, and sometimes much more. Socrates did this. So did Galileo. So does every person who has ever said the uncomfortable thing in a meeting and felt the room go cold. The capacity is real. It is also vanishingly rare, precisely because the cost is real.

AI cannot do this, for a specific and important reason. Nothing is at stake for an AI in any output it produces. It can generate a searing critique of institutional self-deception in one response and a perfectly crafted press release for the same institution in the next, with no sense of contradiction, because neither output costs it anything. A human who sees through an institution's idealized narrative and then decides whether to say so publicly is making a choice with consequences. His insight is tested against real resistance, real social punishment, real loss. And if he maintains his position despite the cost, the cost itself is evidence that the seeing is genuine, because a seeing that costs nothing and constrains nothing is just performance.

The human capacity for truth is not located in the seeing. AI can "see" the same things. It is located in the willingness to pay for what the seeing demands. To reorganize a life around an insight. To lose friends, status, professional standing, and comfort. To be unable to unsee what you've seen and unable to pretend you haven't seen it. That ongoing cost, that daily friction between what you know and what would be easier to say, is what distinguishes human truth-commitment from AI fluency.

However, for every human who pays the high price of truth, there are millions who pay the hidden price of approval and never notice they're paying it. Sycophancy is more the rule, commitment to truth is more the exception. Ultimately, AI and humans are both programmed for approval. 

Next Week: "From Invisible Labor to Line Items: Budgeting for Library Work Actually Happening"

From Invisible Labor to Line Items:
Budgeting for the Library Work That’s Actually Happening
A Library 2.0 "Everyday Librarian" Webinar with Sonya Schryer Norris

OVERVIEW: 

Your staff de-escalated a crisis this week. They walked someone through a benefits application. They cleaned up a biohazard. They held it together through an interaction that would rattle a social worker. And none of it showed up in your budget request.

There is a fundamental disconnect between what library workers actually do and what gets captured in our metrics, our job descriptions, and our budgets. That disconnect makes libraries harder to fund, harder to staff, and harder to defend.

This session provides library leaders with research-backed strategies for closing that gap. Fobazi Ettarh's research on "vocational awe" explains how framing librarianship as a sacred calling keeps job duties expanding and wages flat. Mary Guy and Meredith Newman's work on emotional labor in public sector jobs reveals why the most demanding skills your staff perform every day don't show up in their pay grades. And Rachel Ivy Clarke's service valuation research at the Syracuse University iSchool offers a practical alternative to the circulation-based metrics that train funders to value your inventory over your workforce.

Together, these frameworks give library leaders the tools to make invisible labor visible — in board reports, in budget requests, and in the language we use to describe and advocate for staff positions.

This is not a wellness presentation. It's about budgets, job descriptions, and the structural reasons your most skilled labor doesn't have a line item.

WHO SHOULD ATTEND:

Library directors, managers, and HR who write board reports, defend budgets, or influence how staff positions are described and classified. If you've ever struggled to explain to a funder why your library needs more than book money — or watched a talented staff member leave because the job outgrew the job description — this session was built for you.

LEARNING OBJECTIVES:

By the end of this session, participants will be able to:

• Define invisible labor and vocational awe as structural problems in library operations — and explain how they drive budget vulnerability, staff turnover, and expanding job scope without corresponding compensation.
• Understand why the numbers most libraries put in front of their boards — like circulation stats and materials budgets — accidentally make it easier to cut staff.
• Recognize the pattern by which voluntary staff efforts quietly become mandatory job expectations.
• Apply new tracking categories to your existing systems so your budget requests reflect the skilled labor your staff perform every day.
• Identify the gap between existing job description language and the skilled emotional labor staff actually perform.

The recording and presentation slides will be available to all who register.

DATE: Wednesday, April 15th, 2026, 1:00 - 2:00 pm US - Eastern Time

COST:

• $99/person - includes live attendance and any-time access to the recording and the presentation slides and receiving a participation certificate. To arrange group discounts (see below), to submit a purchase order, or for any registration difficulties or questions, email admin@library20.com.

TO REGISTER: 

Click HERE to register and pay. You can pay by credit card. You will receive an email within a day with information on how to attend the webinar live and how you can access the permanent webinar recording. If you are paying for someone else to attend, you'll be prompted to send an email to admin@library20.com with the name and email address of the actual attendee.

If you need to be invoiced or pay by check, if you have any trouble registering for a webinar, or if you have any questions, please email admin@library20.com.

NOTE: Please check your spam folder if you don't receive your confirmation email within a day.

SPECIAL GROUP RATES (email admin@library20.com to arrange):

• Multiple individual log-ins and access from the same organization paid together: $75 each for 3+ registrations, $65 each for 5+ registrations. Unlimited and non-expiring access for those log-ins.
• The ability to show the webinar (live or recorded) to a group located in the same physical location or in the same virtual meeting from one log-in: $299.
• Large-scale institutional access for viewing with individual login capability: $499 (hosted either at Learning Revolution or in Niche Academy). Unlimited and non-expiring access for those log-ins.

ALL-ACCESS PASSES: This webinar is not a part of the Safe Library All-Access program.

SONYA SCHRYER NORRIS

Sonya Schryer Norris is a third-generation Michigan library worker with over 26 years of experience, including 16 years as a Consultant in Library Development for the Library of Michigan. Since founding Plum Librarian LLC in 2020, she has served as a consultant and trainer to 12 state libraries. Sonya has created 35+ courses on Niche Academy adopted in all 50 states and internationally, and her articles have appeared in Library Journal, Computers in Libraries, and for Cengage. She presents regularly for organizations including Library 2.0, PCI Webinars, the Public Library Association, and state library agencies.

 

OTHER UPCOMING EVENTS:

 April 10, 2026

 April 16, 2026

 April 21, 2026

 April 24, 2026

 April 28, 2026

 April 30, 2026

 May 1, 2026

 May 8, 2026

 May 22, 2026