
The cycle that began with [YOU] on AI asks what AI is doing to the people who use it—not merely to the work they produce. Gene-culture coevolution provides the most precise answer: AI is a new and extraordinarily powerful force acting on the cultural half of an ancient coupled system, capable of transforming culture at unprecedented speed, even of becoming a non-human producer within it. If genes set the slow biases and culture works the fast variations atop them, then AI is a new engine of cultural variation, intervening in the cultural processes that shape human behavior faster and more pervasively than any previous technology.
The framework reveals the danger Wilson named in his most quoted sentence as a structural claim, not a rhetorical flourish. We have Paleolithic emotions, medieval institutions, and godlike technology: genes change at evolutionary speed, essentially not at all on the timescale of technological development. Culture, supercharged by AI, can now change with terrifying rapidity. The result is a widening gap between the genetic substrate, frozen by the slowness of biological evolution, and the cultural environment, accelerating beyond anything our evolved psychology was prepared for.
The framework also introduces a genuinely novel possibility. Wilson's coevolving culture was made and transmitted by human beings; the cultural side of the system was always a human production. But AI now generates culture—text, images, music, ideas, arguments—and increasingly that artificial culture circulates, is consumed by humans, and shapes them, while also serving as training data for the next generation of AI. The cultural partner in the coevolutionary system is beginning to include a non-human producer, and a feedback loop is forming in which AI-generated culture shapes humans and trains further AI, with human production an ever-smaller share. This is a genuinely unprecedented situation that Wilson's framework, designed for the coevolution of human genes and human culture, must now stretch to accommodate.
Wilson developed gene-culture coevolution with Charles Lumsden, a physicist trained in complex systems, whose mathematical background complemented Wilson's biological expertise. Their collaboration produced Genes, Mind, and Culture (1981) and Promethean Fire (1983)—books that attempted to give the interaction between biology and culture a formal mathematical treatment. The core concept is the epigenetic rule: a genetic bias in development that channels the direction of cultural learning without determining it, making certain cultural forms easier to acquire, more stable, and more widely distributed than others.
The framework encountered skepticism from both sides of the nature-nurture debate—too biological for social scientists committed to cultural constructionism, too cultural for behavioral geneticists who preferred simpler models. But its central insight was vindicated by subsequent decades of research in cultural evolution, evolutionary psychology, and cognitive science: human culture does not vary freely over a blank slate, but it is not biologically fixed either. It varies within the constraints and along the channels that evolved psychology provides, and those channels themselves shift, slowly, in response to sustained cultural pressure. The framework is less a theory than a research program, and its program has become central to the fields that have most advanced the understanding of human behavior.
Epigenetic rules. The specific mechanism linking genes to culture: developmental biases that make certain cultural responses easier, more stable, or more widely distributed. Epigenetic rules are not commands but tendencies—they channel cultural learning without determining it, explaining why some cultural universals (basic color categories, incest avoidance, kin-based cooperation) recur across unconnected societies while others vary freely.
The coupled system. Culture and genetics form a single coevolving system, with culture as the faster partner and genetics as the slower. Neither partner is determinative; each constrains and shapes the other across different timescales. This is Wilson's answer to both genetic determinism and cultural constructionism: human nature is real, biological, and constraining—and it is also plastic, variable, and shaped by the cultural environment it helps produce.
The AI disruption. Artificial intelligence is now applying a force of unprecedented magnitude to the fast-moving cultural partner in this coupled system, while the slow-moving genetic partner cannot respond. The result is the widening mismatch Wilson named as the most dangerous feature of the present: Paleolithic emotions encountering godlike technology, with medieval institutions attempting to mediate. Gene-culture coevolution makes this mismatch structurally precise rather than merely rhetorical.
The limits of cultural malleability. Because genes set biases and predispositions, culture cannot make human beings into anything whatever; it works against the grain of an evolved nature that pushes back. AI, however powerfully it reshapes culture, is working on a substrate that resists, and the attempt to use AI to remake human behavior will run into the same evolved bedrock that limited every prior cultural project. The Paleolithic emotions are not infinitely reshapeable even by the most powerful cultural force.
The most contested aspect of gene-culture coevolution is the degree of specificity of the epigenetic rules—how precisely do genes channel cultural learning, and how broadly do they constrain cultural variation? Wilson's critics argued that the rules he identified were too loosely specified to generate precise predictions and that the mathematical formalism in Genes, Mind, and Culture outran the biological evidence. The framework's defenders countered that the critics demanded a precision the empirical material could not yet supply, and that subsequent decades of research in evolutionary psychology and cultural evolution have provided mounting evidence that cultural variation is indeed channeled in the ways the framework predicts. A separate debate concerns the non-human producer problem that AI now raises: Wilson's framework was designed for a system in which both partners—genes and culture—are products of the same biological lineage. AI-generated culture introduces a third term that has no evolutionary history, no biological embeddedness, and no Paleolithic emotional substrate. Whether the coevolutionary framework can accommodate this third term, or whether it requires fundamental revision, is one of the most interesting open questions at the intersection of Wilson's work and the AI transition.