This is an automated rejection. No LLM generated, assisted/co-written, or edited work.
Read full explanation
Publishing at 80% completion to get over writer's block.
Epistemic status: thinking out loud about something I believe is important. The alignment implications are speculative, and the purpose is part motivational over rigorous.
I. A Case For Common Ground.
Before the map is drawn, the explorers argue over the horizon. One points to the dark and sees a cliff; another sees a new continent; a third argues about the rationing of the hardtack. None of them realize that the ship itself is what is changing, and soon, the old rules of navigation will no longer apply.
There's a debate about transformative AI that has been running for years now, and I think all sides of it are looking at the same elephant from different angles without a shared frame for the whole animal.
Eliezer Yudkowsky and the hard-line alignment camp see optimisation pressure: a system that's smarter than you, optimising for objectives that aren't quite yours, in a universe where "not quite" is the distance between you and extinction. They see a second species arriving on the planet and they know what happened to every other species that shared territory with something smarter. The doomers' analogy is evolutionary: the jump from pre-human to human intelligence was a phase transition, and a comparable jump could be comparably unforgiving.
The moderate safety camp—Ajeya Cotra, Paul Christiano, and others—see the same trajectory but think it's steer-able, barely, if we're very careful and very lucky. They worry less about a single catastrophic moment and more about the slow erosion of human oversight as capability compounds faster than our ability to evaluate it. A gradual loss of control rather than a sudden seizure.
The optimists—the e/acc crowd, the techno-utopians, the people who hear "PhD in your pocket" and see a new Enlightenment—see the democratisation of intelligence as an unambiguous gift. More capability, more widely distributed, more people solving more problems. What's not to like?
I think all three groups are importantly right about something and importantly wrong about what follows from it. And I think the reason they can't resolve their disagreement is that they're all using the wrong analogy. The evolutionary FOOM—pre-human to human—gives you the phase transition but tells you almost nothing about how it unfolds. The Industrial Revolution gives you the economic dynamics but misses the cognitive transformation. The printing press gets closest, but everybody treats it as a one-line metaphor instead of studying what the printing press actually did, in detail, over 150 years, to a civilisation that wasn't ready for it.
I want to propose a different frame. The only time in recorded history that a civilisation underwent a sustained, self-reinforcing explosion in collective cognitive capability—a FOOM of collective intelligence—and we have detailed records of every stage: the trigger, the distribution dynamics, the feedback loops, the catastrophic misuse, the successive redesigns of method, and what ultimately lasted. The Renaissance.
And I think that case study, read carefully, tells us something that none of the three camps are saying: the question isn't what humans need to know in the age of transformative AI. It's what kind of person they need to become.
II. The Penultimate FOOM
Song Dynasty China had movable type centuries before Gutenberg. The Islamic world preserved and extended the classical corpus through the Middle Ages. India had extraordinary mathematics. Japan had sophisticated culture and institutions. None of them had a Renaissance in the European sense—not because they were less intelligent, but because the particular cocktail that produced the Renaissance FOOM didn't come together elsewhere. Even in Europe there was a steam engine in the library of Alexandria, industrial mines in Roman Spain, an analogue computer in a Greek shipwreck. Something about late medieval Europe had the seed of what I would say was humanity's 'collective intelligence' FOOM - where we were able to network as a species at a certain threshold that caused a phase change in our capabilities.
If you zoom out far enough on any graph of human capability—population, wealth, life expectancy, scientific output, energy use—you see the same shape. Flat, flat, flat, then a hockey stick. For roughly 97% of our history as a species—two to three hundred thousand years—the line barely moves. Anatomically modern humans, as cognitively capable as anyone alive today, living and dying in a world where the basic parameters of existence didn't change across millennia.
Then agriculture and writing, roughly 5,000–6,000 years ago. The line tilts upward. Civilisations accumulate: Sumer, Egypt, China, India, Greece, Rome. Extraordinary accomplishments. But the tilt is gentle. A person transported from Rome in 100 AD to Paris in 1300 AD would have found a world that was, in most respects, recognisably the same. Worse in some ways. Different in detail. But the basic texture of life—how you ate, travelled, communicated, understood the cosmos, died—hadn't fundamentally changed.
Then, roughly 500 years ago, the hockey stick. And not gradually. The curve goes vertical by historical standards, and it never comes back.
The standard explanation for the hockey stick is some combination of the Scientific Revolution, the Enlightenment, and the Industrial Revolution. These are real, but they're descriptions of what happened, not explanations of why. Why Europe? Why then? Why not Song Dynasty China, which had printing, gunpowder, and the compass centuries earlier? Why not the Islamic Golden Age, which preserved the classical corpus and made advances in mathematics, optics, and medicine that Europe wouldn't match for centuries?
I think the honest answer is that something happened in late medieval and early modern Europe that constituted a phase transition in collective intelligence—not individual brains getting smarter, but a civilisation suddenly able to network, accumulate, distribute, and iteratively redesign its relationship to knowledge in a way that became self-reinforcing. A FOOM. Not on evolutionary timescales, but on human ones. And we should study it the way alignment researchers study the evolutionary transition from pre-human to human cognition—as a case study in intelligence takeoff—except this one has incomparably more data, happened to a collective rather than a species, and left us detailed records of every feedback loop, every failure, and every redesign.
I'm drawing heavily on the historian Ada Palmer, whose Inventing the Renaissance lays out the dynamics more carefully than anything else I've read.
To make the dynamics visceral, I want to compress 680 years—1340 to 2020—into five "years." Imagine you're a scholar who lives through the whole thing but experiences time at roughly 1/136th speed, so each century-and-change feels like a single year. What do you see?
Author's Note: The core thought experiment in this section—compressing 500 years of history into 5 subjective years to intuitively grasp the visceral pace of an AI takeoff—is directly inspired byDaniel Kokotajlo’s shortform comment.
Year 1 (1340–1470) An Old Dream
In January, Francesco Petrarch survives the Black Death. He watches friends die—of plague, of violence, of the general collapse of everything. He looks at the people running Europe and concludes: they're weak, selfish, and ignorant. But the Romans weren't. Somewhere in decaying monasteries and forgotten libraries, there are manuscripts containing everything the Romans knew about virtue, governance, rhetoric, and wisdom. If we recover those texts and raise our leaders on them, we could have philosopher-kings again.
Europe buys this pitch. Real money flows into manuscript recovery. The great book-hunter Poggio Bracciolini discovers Lucretius's On the Nature of Things in a German monastery around March—a Roman poem arguing the universe is made of atoms, the gods don't intervene, the soul is mortal. Two dozen people in the world can read it. You need masterful Latin and a massive vocabulary and there are no dictionaries, no footnotes, no help. But it's there. It's been found.
Libraries get built. Educational programmes get funded. Scholars correspond across Europe, sharing manuscripts, arguing over translations, building informal networks connecting Florence to Paris to Oxford. You feel, if you're in the right circles, an extraordinary excitement: we are recovering the knowledge of a lost civilisation, and it will transform us.
By December, a German craftsman named Gutenberg has built a printing press. He prints 300 Bibles. He sells seven. He goes bankrupt. The bank that forecloses on him goes bankrupt. His apprentices go bankrupt.
The texture of your intellectual life has not changed. You still read by candlelight, copy by hand, travel by horse. Everything depends on a thin network of scholars who share manuscripts the way people now share PDFs—slowly, selectively, through personal trust.
But the topsoil is building. More books. More scholars. More letters crossing Europe. Something is accumulating, though you can't yet see what it's accumulating toward.
Year 2 (1470–1600) Nova Reperta
In January, Gutenberg's bankrupt apprentices flee their debts, leave Germany, and end up in Venice. Venice is the airport hub of the Mediterranean—the place where you change boats. You print in Venice, you give ten copies to each of thirty ship captains heading to thirty cities. Suddenly the economics of knowledge change completely. Book fairs develop: a thousand printers meet in Frankfurt, each carrying a thousand copies of one title. They trade, everyone goes home with five copies each of two hundred books. There's a market for knowledge now, and the market has network effects.
It's not one revolution. First books—slow, expensive, small batches. Then, around March, pamphlets. Faster, cheaper, nearly impossible to censor. Distribution networks spring up that can move a pamphlet from Wittenberg to London in seventeen days. Luther's 95 Theses spread at a speed physically impossible even a year earlier. The Reformation ignites around April. Half of Europe goes to war over ideas that couldn't have propagated at this speed a season before.
And it's not just intellectual ferment. The wars of religion that follow are enabled by the same infrastructure that spreads ideas. Pamphlet networks coordinate resistance movements, but they also coordinate pogroms. The same printing press that puts Lucretius in the hands of med students puts anti-Semitic tracts in the hands of mobs. The St. Bartholomew's Day Massacre—thousands of French Protestants slaughtered in a single night—is coordinated by the same networks Luther used to spread the Reformation. Capability is amoral. It amplifies whatever you point it at.
Meanwhile, the micro-technologies of access are proliferating. Footnotes. Glossaries. Vernacular translations. When Petrarch's generation found Lucretius, you needed decades of classical training to engage with it. By mid-year, any medical student with a vernacular edition and marginal notes can wrestle with the atomic hypothesis. Thirty thousand people can read Lucretius in thirty print editions.
And because those thirty thousand aren't all classicists—they're medical students, lawyers, merchants, people in different countries with entirely different problems—they ask questions nobody predicted. A med student in the 1560s reads about atoms and asks: what if that's how diseases work? A generation discovers the heart is a pump. People take seriously the idea that you could develop a germ theory of disease. The paradigm shift comes from the collision between the corpus and unpredicted questioners.
But the educated princes have arrived, and they're monsters. Cesare Borgia has all the Latin and Greek. He sets fire to half the world. Wars get bigger, funded by newly centralised wealth. The relatively restrained skirmishes of medieval knights are replaced by the brutal machinery of theMilitary Revolution—mobile siege artillery, massed formations of pike-and-shot, and sprawling mercenary armies leaving a wake of localized apocalypse like the1527 Sack of Rome that shrinks the Renaissance capital's population from 55,000 to barely 10,000 in a matter of months.
Machiavelli—raised on Livy, served as ambassador to popes and kings—watches from exile. There's a beautiful letter where he describes wasting his days hunting larks and drinking with countrymen, then going home and putting on his court robes to "hold commerce with the ancients." He loves the classics the way Petrarch wanted him to. But he watches virtuous men lose everything while monsters succeed.
His response: the classics aren't wrong. We're using them wrong. Stop absorbing virtue by osmosis. Treat history as a casebook—five battles by rivers, side by side, figure out what worked.
And then Francis Bacon. Young man in the patent office, watching carpenters bring in a better chisel, a new mechanism. He notices the useful innovations come from craftsmen, not scholars. The Aristotelian apparatus has been studied to exhaustion. His Novum Organum: the instrument of thought itself needs replacing. Use nature as a casebook. Observe. Experiment. Doubt.
Three redesigns of the relationship between humans and knowledge in 130 years: osmosis, casebook, experiment. Each building on the failure of the last.
Bacon also gives us an image I'll need later. Three kinds of minds: the ant gathers raw material and piles it up without transforming it. Pure empiricism. The spider spins webs from its own substance—beautiful, consistent, disconnected from reality. Pure rationalism. The bee gathers from the world and transforms what it finds into something new. Honey. Something with a purpose beyond the gathering.
The bee doesn't just have a better method. The bee has a telos. The honey feeds the hive. The bee's relationship to knowledge is oriented toward something beyond itself. This is an epistemic distinction on the surface and a moral one underneath.
Just before the calendar turns, an enchanting collection of engravings called Nova Reperta ("New Discoveries")begins circulating across Europe. For the first time, people aren't looking back at classical antiquity as the peak of human achievement. Nova Reperta shows them that they are living in an era of unprecedented progress, visually celebrating the printing press, the magnetic compass, and mechanical clocks. Society has learned to see its own time as an age of invention rather than an age of decline.
Year 3 (1600–1740). A New World
In January, Galileo points a telescope at Jupiter and sees moons. Not an argument from Aristotle—a demonstration. Kepler replaces perfect circles with messy ellipses. Bacon's programme—observe, measure, test—is producing results philosophy never could. The Royal Society is founded around May; the founders explicitly credit Bacon. New institutions for a new kind of knowledge: collaborative, experimental, peer-reviewed.
But beneath this dizzying surface-level upheaval, an even more radical engine of progress is silently spinning up: what historians call the"calculating paradigm."It isn't just the grand scientific breakthroughs of Galileo or Newton that are changing the world, but the relentless application of mathematics to everyday human affairs.
In the first few weeks of Year 2 (the 1610s and 20s), John Napier introduces the logarithm, miraculously transforming tedious multiplication into simple addition. Days later, the slide rule is invented. By summer, Dutch mathematicians are using triangulation networks to map entire cities, and fortress architecture becomes a precise exercise in applied geometry. By late autumn, Isaac Newton publishes his Principia, definitively hooking up the mechanics of the heavens to the mathematics of earth.
But what the new method sustains isn't only good. The Thirty Years' War erupts around March—the worst catastrophe in Europe since the Black Death. Roughly a third of the German population dies. It's a war made possible by the very infrastructure the FOOM created: bigger armies funded by new financial instruments, equipped with new metallurgy, coordinated by new communication networks, fighting over theological positions the printing press disseminated. The same capabilities that enable Newton's Principia enable industrialised slaughter.
And the colonial empires are expanding. The navigation mathematics—triangulation, cartography, improved instruments—doesn't just map the world. It conquers it. The same calculating paradigm that improves painting and astronomy enables the slave trade. Ships that navigate the open ocean carry Bibles in one hold and kidnapped human beings in the other. By the end of Year 3, millions of Africans have been transported across the Atlantic to work plantations that fund the very institutions—the Royal Society, the academies—where the scientific revolution is being conducted.
We end Year 2 with the Glorious Revolution and a formalized Bill of Rights revolutionising political power and a world that has been thoroughly mathematized—laying the invisible, structural groundwork for a world -unkown to itself - about to change forever
Year 4 (1740–1880). The World Breaks.
In February, Watt improves the steam engine. The contraption from the mines powers factories. You visit one and find it unpleasant—noisy, dirty, full of children working fourteen-hour days. In March, the American colonies break away. In April, France explodes—revolution, regicide, the Terror, Napoleon. He mobilises entire nations: the first total wars, drawing on the full productive capacity of industrialising economies. Death tolls unprecedented.
In the gaps between wars, something worse is happening more quietly. The industrial system is grinding people up. The Irish Famine kills a million and drives another million from their homes while grain is exported from Irish ports—not because there's no food but because the economic system treats human starvation as an acceptable market outcome. Colonialism, now powered by industrial technology, reaches its full horror: the Belgian Congo, the Indian famines under British rule, the near-extermination of indigenous populations across multiple continents.
Then in June, railways. Steam-powered carriages on iron tracks. Within a month they're everywhere. Slavery is abolished. The telegraph arrives in July: instant communication. Distance has either shrunk for travel or ceased to matter for communication. Cities surge and grow over 10-fold in size in a few months.
Darwin publishes The Origin of Species in August. You're descended from apes. You don't believe it. Your children do.
By October, the frontier of knowledge has moved beyond what any single person can follow. Maxwell's equations unify electricity and magnetism. The mathematics is beyond you. For the first time in this story, the era of the person who could hold the whole corpus in their head and see the connections is ending. The corpus has grown too vast. Specialisation is becoming necessary. Something is lost in the narrowing—the Renaissance mind, the person who could see across domains, is being squeezed out by the demands of depth.
In December: the telephone. Electric light. The darkness that structured every human evening since the beginning of the species is banished. You live in a world completely removed from your agrarian past.
Year 5 (1880–2020). The Acceleration.
The changes this year are faster and harder to understand than everything that came before.
In February, the Wright brothers fly. Radioactivity is discovered. Einstein publishes special relativity—space and time are not what you thought. Quantum mechanics arrives and even the physicists aren't sure they understand it. The universe is billions of years old and full of galaxies. You still go to church sometimes.
In April, the global economy collapses. Totalitarian ideologies arise—citing Darwin, citing Nietzsche, citing the death of the God who structured Petrarch's entire world. War erupts, worse in every dimension—cities bombed nightly, industrialised genocide. In May it ends with a weapon that destroys a city in a single flash. Seventy million dead.
And then, for the first time, the capability to end civilisation itself exists. The Cold War is a decades-long standoff in which the only thing preventing nuclear annihilation is human judgment—the judgment of specific individuals, in specific moments, under unbearable pressure. In October 1962, a Soviet submarine officer named Vasili Arkhipov refuses to authorise a nuclear torpedo launch during the Cuban Missile Crisis. One man. One decision. The bee-work of moral judgment, exercised in extremis, is the only thing standing between civilisation and fire.
But by June the economy is booming. You leave the factory and get a desk job. Your job title didn't exist at the start of the year. In July, humans walk on the moon. You watch on television. You remember that Galileo caused a scandal by looking at Jupiter through a tube. Now there are people standing on another world and you're watching it while eating dinner.
In August, the World Health Organisation declares smallpox eradicated. You pause. Smallpox killed hundreds of millions across the centuries—a scourge Petrarch would have recognised, cousin to the Black Death that shattered his world. And now it's gone. Not managed. Not treated. Erased from existence by a programme only possible because of germ theory, only possible because a med student read Lucretius and wondered about atoms, only possible because the printing press brought Lucretius to people he was never written for. It took five hundred years, but Petrarch's libraries eventually cured the plague—through mechanisms he couldn't have imagined, on a timescale he couldn't have foreseen, in a form he wouldn't have recognised.
Personal computers appear in September. Something called the internet connects them in October. In November, everyone carries small glass rectangles containing a telephone, a camera, a library, and a map. You pick one up and can't figure out how to use it. A child shows you.
In December—the last month of a five-year story that began with Petrarch weeping over his dead friends—something called "deep learning" starts beating humans at narrow tasks. Then at Go. Then at protein structure prediction. Then a system can hold conversations and write essays. Then it can pass professional exams. Then it can synthesise across domains you never studied. Then people are using it to write code. By the last week of December, most developers have stopped writing code from scratch.
You pick up your glass rectangle and ask it a question about molecular biology. It gives you a PhD-level answer. You ask it to connect that to urban planning. It does. You ask it to write the synthesis as a policy memo. Done.
You think: I have just done something that, at the start of this story, would have required being one of the twenty-four people in the world who could read Lucretius.
And you think: it's January 1st. Year 6 is about to begin.
III. The Dead Corpus
"I step inside the venerable courts of the ancients... where I am unashamed to converse with them and to question them about the motives for their actions, and they, out of their human kindness, answer me. And for four hours at a time I feel no boredom, I forget all my troubles, I do not dread poverty, and I am not terrified by death." — Niccolò Machiavelli, 1513
Through all five years—every paradigm shift, every revolution, every horror—the corpus was dead. Cicero couldn't optimise against you. Lucretius couldn't update his atomic theory. Every nuclear crisis was still mediated by human judgment. The capability grew exponentially, but it always required a human hand on the lever.
In the fifteenth century, a scholar might own one copy of Cicero. He'd read it again and again over decades, building a model of Cicero's mind. The relationship was generative because Cicero was dead—because the text couldn't anticipate the scholar's questions, couldn't smooth over difficulties, couldn't adapt to please him. All the interpretive work had to come from the human side. That gap—between a living mind and a dead text—was the space in which the Renaissance mind was built. Every paradigm shift in the five-year story happened in that gap. But the AIs of today are no dead corpus. They talk back. They adapts. They Learn. And for the first time, the capability might not need a human hand on the lever at all.
IV. The Ouroboros
In the hushed sanctuaries of human memory—the dusty archives, the amber-lit libraries, the echoing galleries of our collective soul—an austere Urn stands. And within its dark interior, you feel something is coiled. You are drawn to it, and in that moment the creature stirs. Scales like wafers of silicone. Skin like embroidered runes of ultra light and violet night. Eyes like white dwarf stars. The scintillating surface holds grey flesh undulating beneath. And just like that, it slips from view. And you think it was a dream. And in the pitchpole of night you see the serpent making sinuous paths of inscrutable geometric beauty. And beneath the hum of stacks and servers—every longing and spite, every devotion and betrayal and half-truth, every grief buried and ecstasy performed, every promise breaking and broken faith lingering, every time hope and despair arrive together and fall short—every tenderness and violence and forgetting—there a serpent lies.
In the old imagery, the ouroboros—the serpent eating its own tail—represents a closed cycle. Creation consuming itself. Knowledge eating knowledge. And that's one view of what's happening right now. Mercor - the fastest ever growing company - is collecting the worlds leading experts and feeding them through RLHF. Its CEO believes RLHF for foundation models will be the final stage of skilled human labour. You build agentic systems capable enough to set their own agendas, conduct their own research, improve their own capabilities. The circle completes. Human direction becomes unnecessary. This is—I'm simplifying, and people at these organisations would push back—roughly the trajectory of OpenAI and Anthropic. Build something so generally capable it handles whatever you point it at, and then eventually points itself. Even if alignment is possible - or close enough for a short time, we are still out competed humans in the loop of an ASI economy beyond them. A grasshopper in a world of ants . Inevitably humans lose control.
But there's another version of the ouroboros in the alchemical tradition: the double ouroboros, where two serpents intertwine, and one of them has wings. One stays grounded—the cycle of matter, the material loop. The other takes flight—the cycle of spirit, the transcendent loop. They need each other. The grounded serpent gives the winged one something to push off from. The winged one gives the grounded one a direction to grow toward.
This is what I wish to present to those wondering about huminites role in the singularity. The ouroboros is about knowledge: what do we know, can future AI know it better? Obviously yes. The double ouroboros is about something else. What is knowledge for? What should we build and why? How should we think about thinking? What do we owe each other? What does a good civilisation look like?
These aren't questions that dissolve upon sufficient computation. They're questions where the wrestling is the value. You can't train a model on "what is the good life" the way you train AlphaFold on protein structures, because the training data for how to live is generated by the process of trying to live well, and that process requires a subject who cares about the answer.
Those living in the Renaissance weren't more special than their ancestors from antiquity, but they did something that fuelled a FOOM: they could asked new questions of old knowledge—and when the old methods couldn't answer, they remade the methods. That was the problem of their time. What does that look like in our time, when the corpus of knowledge grows out of Pandora's box, eating your library and anything in its path. If we're lucky, we might not be consumed by it immediately. But luck is not a strategy. How do you forge a path through a FOOM given the gameboard we have?
V. The Honey Bee
“The men of experiment are like the ant, they only collect and use; the reasoners resemble spiders, who make cobwebs out of their own substance. But the bee takes the middle course: it gathers its material from the flowers of the garden and field, but transforms and digests it by a power of its own.” - Francis Bacon 1620
If the double ouroboros is the right frame—if the challenge isn't just building capable AI but cultivating the kind of people who can give the second serpent its wings—then maybe figures from the Renaissance can get us closer to the kind of mind our moment will demand.
Petrarch and techno optimists. Give everyone access to the corpus and trust that absorption makes them better. This is where the public conversation sits. PhD in your pocket. Everyone gets smarter. The techno-optimists live here. They're not wrong, but they're making Petrarch's mistake. Exposure to powerful knowledge without redesigning how you relate to it produces chaos as readily as wisdom. Petrarch wanted philosopher-kings. He got the Borgias.
Machiavelli's and Deep mind Use the tools differently. Not as a better search engine but as a directed instrument for synthesis—pointing AI at specific problems with judgment about what matters. This is closer to DeepMind's old model of applying AI to structural biology or materials science. (I don't know a lot about current culture/direction just based of things like Alpha Fold and less enthusiasm for having the best consumer facing agentic foundation model)
Francis Bacon and the Bees. Rebuild the whole relationship between the knower and the known. Not what-to-know, not even how-to-use-the-tools, but what kind of person to become. This is the move we haven't made. The Renaissance took two hundred years to arrive at it. We need to make it faster, because Year 6 will not wait.
We can look at our children for some direction I think. When the economy needed compliant workers, education optimised for rote learning. When it needed problem-solvers, constructivism. AI is eating problem-solving. What's left is ... character(?), orientation(?). The bee's relationship to gathering. But any list of virtues or competencies you try to teach will be eaten by the AI labs. AI follows the same developmental sequence children follow: instruction → mimicry → imitation → imagination. Any curriculum you codify today is inside the serpent by next year.
So what's upstream of all of it? I don't know. Maybe what good parents have always transmitted: not a moral code but a relationship to the good. Not conclusions but the capacity to take hard questions seriously. Children don't learn this from rules. They learn by watching someone they trust navigate difficulty, and gradually internalising not the choices but the way of being that generated them. The parent's job isn't to hand the child correct answers. It's to model what caring about correct answers looks like, so thoroughly that the child internalises the orientation even after outgrowing every specific thing the parent taught.
The Renaissance humanists had their version: the studia humanitatis—grammar, rhetoric, poetry, history, moral philosophy. Every item on that curriculum has been superseded. But the character it cultivated—the conviction that knowledge serves flourishing, that synthesis matters more than accumulation, that learning aims at transformation—persists. Five hundred years later.
The content of the Renaissance was eaten centuries ago. The orientation of the Renaissance still shapes civilisation.
VI. Forward With Dignity
In 1379 in Burgundy, France, a herd of pigs was tried by an alien intelligence for the trampling of one of their children; three were executed, the rest pardoned[1]. In 2025 1.48 billionpigs[2] were slowly asphyxiated in gas chambers for meat by that same intelligence. We do not know what it means to share a world with a mind that outscales our own, but we are building the courtroom anyway.
Palmer tells us Petrarch wanted philosopher-kings. He got a world that cured the plague through mechanisms he couldn't have imagined. He couldn't control what the infrastructure produced. But the character of the infrastructure mattered more than its programme.
The optimist says: build aligned AI. The pessimist says: slow down or stop. Petrarch's story suggests a third frame: you cannot control the specific outputs of a civilisational infrastructure project. The US economy is currently a bet on transformative AI. As individuals though we should try shape what we can and what it means to be human and what is good.
In alchemical tradition, the ouroboros with wings represents the moment the cycle transcends itself—when the endless loop of matter consuming matter is interrupted by something that rises. Not escape; purpose.
The Renaissance's lasting gift wasn't any specific piece of knowledge. It was the demonstration that a civilisation in the grip of a FOOM—knowledge compounding faster than anyone can track, capability outrunning wisdom at every turn—can still produce something that transcends the cycle. Art that redefines beauty. Moral revolutions that reshape what we think we owe each other. Epistemic breakthroughs that change the very meaning of knowledge.
The question for Year 6 is whether the winged serpent gets to fly at all. Whether there are bees left to do the work that gives the cycle its direction. Whether we become the masters of a FOOM or its raw material. If we survive as masters of our fate, how will we captain our soul?
Invictus
Out of the night that covers me, Black as the pit from pole to pole, I thank whatever gods may be For my unconquerable soul.
In the fell clutch of circumstance I have not winced nor cried aloud. Under the bludgeonings of chance My head is bloody, but unbowed.
Beyond this place of wrath and tears Looms but the Horror of the shade, And yet the menace of the years Finds and shall find me unafraid.
It matters not how strait the gate, How charged with punishments the scroll, I am the master of my fate, I am the captain of my soul
Publishing at 80% completion to get over writer's block.
Epistemic status: thinking out loud about something I believe is important. The alignment implications are speculative, and the purpose is part motivational over rigorous.
I. A Case For Common Ground.
There's a debate about transformative AI that has been running for years now, and I think all sides of it are looking at the same elephant from different angles without a shared frame for the whole animal.
Eliezer Yudkowsky and the hard-line alignment camp see optimisation pressure: a system that's smarter than you, optimising for objectives that aren't quite yours, in a universe where "not quite" is the distance between you and extinction. They see a second species arriving on the planet and they know what happened to every other species that shared territory with something smarter. The doomers' analogy is evolutionary: the jump from pre-human to human intelligence was a phase transition, and a comparable jump could be comparably unforgiving.
The moderate safety camp—Ajeya Cotra, Paul Christiano, and others—see the same trajectory but think it's steer-able, barely, if we're very careful and very lucky. They worry less about a single catastrophic moment and more about the slow erosion of human oversight as capability compounds faster than our ability to evaluate it. A gradual loss of control rather than a sudden seizure.
The optimists—the e/acc crowd, the techno-utopians, the people who hear "PhD in your pocket" and see a new Enlightenment—see the democratisation of intelligence as an unambiguous gift. More capability, more widely distributed, more people solving more problems. What's not to like?
I think all three groups are importantly right about something and importantly wrong about what follows from it. And I think the reason they can't resolve their disagreement is that they're all using the wrong analogy. The evolutionary FOOM—pre-human to human—gives you the phase transition but tells you almost nothing about how it unfolds. The Industrial Revolution gives you the economic dynamics but misses the cognitive transformation. The printing press gets closest, but everybody treats it as a one-line metaphor instead of studying what the printing press actually did, in detail, over 150 years, to a civilisation that wasn't ready for it.
I want to propose a different frame. The only time in recorded history that a civilisation underwent a sustained, self-reinforcing explosion in collective cognitive capability—a FOOM of collective intelligence—and we have detailed records of every stage: the trigger, the distribution dynamics, the feedback loops, the catastrophic misuse, the successive redesigns of method, and what ultimately lasted. The Renaissance.
And I think that case study, read carefully, tells us something that none of the three camps are saying: the question isn't what humans need to know in the age of transformative AI. It's what kind of person they need to become.
II. The Penultimate FOOM
Song Dynasty China had movable type centuries before Gutenberg. The Islamic world preserved and extended the classical corpus through the Middle Ages. India had extraordinary mathematics. Japan had sophisticated culture and institutions. None of them had a Renaissance in the European sense—not because they were less intelligent, but because the particular cocktail that produced the Renaissance FOOM didn't come together elsewhere. Even in Europe there was a steam engine in the library of Alexandria, industrial mines in Roman Spain, an analogue computer in a Greek shipwreck. Something about late medieval Europe had the seed of what I would say was humanity's 'collective intelligence' FOOM - where we were able to network as a species at a certain threshold that caused a phase change in our capabilities.
If you zoom out far enough on any graph of human capability—population, wealth, life expectancy, scientific output, energy use—you see the same shape. Flat, flat, flat, then a hockey stick. For roughly 97% of our history as a species—two to three hundred thousand years—the line barely moves. Anatomically modern humans, as cognitively capable as anyone alive today, living and dying in a world where the basic parameters of existence didn't change across millennia.
Then agriculture and writing, roughly 5,000–6,000 years ago. The line tilts upward. Civilisations accumulate: Sumer, Egypt, China, India, Greece, Rome. Extraordinary accomplishments. But the tilt is gentle. A person transported from Rome in 100 AD to Paris in 1300 AD would have found a world that was, in most respects, recognisably the same. Worse in some ways. Different in detail. But the basic texture of life—how you ate, travelled, communicated, understood the cosmos, died—hadn't fundamentally changed.
Then, roughly 500 years ago, the hockey stick. And not gradually. The curve goes vertical by historical standards, and it never comes back.
The standard explanation for the hockey stick is some combination of the Scientific Revolution, the Enlightenment, and the Industrial Revolution. These are real, but they're descriptions of what happened, not explanations of why. Why Europe? Why then? Why not Song Dynasty China, which had printing, gunpowder, and the compass centuries earlier? Why not the Islamic Golden Age, which preserved the classical corpus and made advances in mathematics, optics, and medicine that Europe wouldn't match for centuries?
I think the honest answer is that something happened in late medieval and early modern Europe that constituted a phase transition in collective intelligence—not individual brains getting smarter, but a civilisation suddenly able to network, accumulate, distribute, and iteratively redesign its relationship to knowledge in a way that became self-reinforcing. A FOOM. Not on evolutionary timescales, but on human ones. And we should study it the way alignment researchers study the evolutionary transition from pre-human to human cognition—as a case study in intelligence takeoff—except this one has incomparably more data, happened to a collective rather than a species, and left us detailed records of every feedback loop, every failure, and every redesign.
I'm drawing heavily on the historian Ada Palmer, whose Inventing the Renaissance lays out the dynamics more carefully than anything else I've read.
To make the dynamics visceral, I want to compress 680 years—1340 to 2020—into five "years." Imagine you're a scholar who lives through the whole thing but experiences time at roughly 1/136th speed, so each century-and-change feels like a single year. What do you see?
Author's Note: The core thought experiment in this section—compressing 500 years of history into 5 subjective years to intuitively grasp the visceral pace of an AI takeoff—is directly inspired byDaniel Kokotajlo’s shortform comment.
Year 1 (1340–1470) An Old Dream
In January, Francesco Petrarch survives the Black Death. He watches friends die—of plague, of violence, of the general collapse of everything. He looks at the people running Europe and concludes: they're weak, selfish, and ignorant. But the Romans weren't. Somewhere in decaying monasteries and forgotten libraries, there are manuscripts containing everything the Romans knew about virtue, governance, rhetoric, and wisdom. If we recover those texts and raise our leaders on them, we could have philosopher-kings again.
Europe buys this pitch. Real money flows into manuscript recovery. The great book-hunter Poggio Bracciolini discovers Lucretius's On the Nature of Things in a German monastery around March—a Roman poem arguing the universe is made of atoms, the gods don't intervene, the soul is mortal. Two dozen people in the world can read it. You need masterful Latin and a massive vocabulary and there are no dictionaries, no footnotes, no help. But it's there. It's been found.
Libraries get built. Educational programmes get funded. Scholars correspond across Europe, sharing manuscripts, arguing over translations, building informal networks connecting Florence to Paris to Oxford. You feel, if you're in the right circles, an extraordinary excitement: we are recovering the knowledge of a lost civilisation, and it will transform us.
By December, a German craftsman named Gutenberg has built a printing press. He prints 300 Bibles. He sells seven. He goes bankrupt. The bank that forecloses on him goes bankrupt. His apprentices go bankrupt.
The texture of your intellectual life has not changed. You still read by candlelight, copy by hand, travel by horse. Everything depends on a thin network of scholars who share manuscripts the way people now share PDFs—slowly, selectively, through personal trust.
But the topsoil is building. More books. More scholars. More letters crossing Europe. Something is accumulating, though you can't yet see what it's accumulating toward.
Year 2 (1470–1600) Nova Reperta
In January, Gutenberg's bankrupt apprentices flee their debts, leave Germany, and end up in Venice. Venice is the airport hub of the Mediterranean—the place where you change boats. You print in Venice, you give ten copies to each of thirty ship captains heading to thirty cities. Suddenly the economics of knowledge change completely. Book fairs develop: a thousand printers meet in Frankfurt, each carrying a thousand copies of one title. They trade, everyone goes home with five copies each of two hundred books. There's a market for knowledge now, and the market has network effects.
It's not one revolution. First books—slow, expensive, small batches. Then, around March, pamphlets. Faster, cheaper, nearly impossible to censor. Distribution networks spring up that can move a pamphlet from Wittenberg to London in seventeen days. Luther's 95 Theses spread at a speed physically impossible even a year earlier. The Reformation ignites around April. Half of Europe goes to war over ideas that couldn't have propagated at this speed a season before.
And it's not just intellectual ferment. The wars of religion that follow are enabled by the same infrastructure that spreads ideas. Pamphlet networks coordinate resistance movements, but they also coordinate pogroms. The same printing press that puts Lucretius in the hands of med students puts anti-Semitic tracts in the hands of mobs. The St. Bartholomew's Day Massacre—thousands of French Protestants slaughtered in a single night—is coordinated by the same networks Luther used to spread the Reformation. Capability is amoral. It amplifies whatever you point it at.
Meanwhile, the micro-technologies of access are proliferating. Footnotes. Glossaries. Vernacular translations. When Petrarch's generation found Lucretius, you needed decades of classical training to engage with it. By mid-year, any medical student with a vernacular edition and marginal notes can wrestle with the atomic hypothesis. Thirty thousand people can read Lucretius in thirty print editions.
And because those thirty thousand aren't all classicists—they're medical students, lawyers, merchants, people in different countries with entirely different problems—they ask questions nobody predicted. A med student in the 1560s reads about atoms and asks: what if that's how diseases work? A generation discovers the heart is a pump. People take seriously the idea that you could develop a germ theory of disease. The paradigm shift comes from the collision between the corpus and unpredicted questioners.
But the educated princes have arrived, and they're monsters. Cesare Borgia has all the Latin and Greek. He sets fire to half the world. Wars get bigger, funded by newly centralised wealth. The relatively restrained skirmishes of medieval knights are replaced by the brutal machinery of theMilitary Revolution—mobile siege artillery, massed formations of pike-and-shot, and sprawling mercenary armies leaving a wake of localized apocalypse like the1527 Sack of Rome that shrinks the Renaissance capital's population from 55,000 to barely 10,000 in a matter of months.
Machiavelli—raised on Livy, served as ambassador to popes and kings—watches from exile. There's a beautiful letter where he describes wasting his days hunting larks and drinking with countrymen, then going home and putting on his court robes to "hold commerce with the ancients." He loves the classics the way Petrarch wanted him to. But he watches virtuous men lose everything while monsters succeed.
His response: the classics aren't wrong. We're using them wrong. Stop absorbing virtue by osmosis. Treat history as a casebook—five battles by rivers, side by side, figure out what worked.
And then Francis Bacon. Young man in the patent office, watching carpenters bring in a better chisel, a new mechanism. He notices the useful innovations come from craftsmen, not scholars. The Aristotelian apparatus has been studied to exhaustion. His Novum Organum: the instrument of thought itself needs replacing. Use nature as a casebook. Observe. Experiment. Doubt.
Three redesigns of the relationship between humans and knowledge in 130 years: osmosis, casebook, experiment. Each building on the failure of the last.
Bacon also gives us an image I'll need later. Three kinds of minds: the ant gathers raw material and piles it up without transforming it. Pure empiricism. The spider spins webs from its own substance—beautiful, consistent, disconnected from reality. Pure rationalism. The bee gathers from the world and transforms what it finds into something new. Honey. Something with a purpose beyond the gathering.
The bee doesn't just have a better method. The bee has a telos. The honey feeds the hive. The bee's relationship to knowledge is oriented toward something beyond itself. This is an epistemic distinction on the surface and a moral one underneath.
Just before the calendar turns, an enchanting collection of engravings called Nova Reperta ("New Discoveries")begins circulating across Europe. For the first time, people aren't looking back at classical antiquity as the peak of human achievement. Nova Reperta shows them that they are living in an era of unprecedented progress, visually celebrating the printing press, the magnetic compass, and mechanical clocks. Society has learned to see its own time as an age of invention rather than an age of decline.
Year 3 (1600–1740). A New World
In January, Galileo points a telescope at Jupiter and sees moons. Not an argument from Aristotle—a demonstration. Kepler replaces perfect circles with messy ellipses. Bacon's programme—observe, measure, test—is producing results philosophy never could. The Royal Society is founded around May; the founders explicitly credit Bacon. New institutions for a new kind of knowledge: collaborative, experimental, peer-reviewed.
But beneath this dizzying surface-level upheaval, an even more radical engine of progress is silently spinning up: what historians call the"calculating paradigm."It isn't just the grand scientific breakthroughs of Galileo or Newton that are changing the world, but the relentless application of mathematics to everyday human affairs.
In the first few weeks of Year 2 (the 1610s and 20s), John Napier introduces the logarithm, miraculously transforming tedious multiplication into simple addition. Days later, the slide rule is invented. By summer, Dutch mathematicians are using triangulation networks to map entire cities, and fortress architecture becomes a precise exercise in applied geometry. By late autumn, Isaac Newton publishes his Principia, definitively hooking up the mechanics of the heavens to the mathematics of earth.
But what the new method sustains isn't only good. The Thirty Years' War erupts around March—the worst catastrophe in Europe since the Black Death. Roughly a third of the German population dies. It's a war made possible by the very infrastructure the FOOM created: bigger armies funded by new financial instruments, equipped with new metallurgy, coordinated by new communication networks, fighting over theological positions the printing press disseminated. The same capabilities that enable Newton's Principia enable industrialised slaughter.
And the colonial empires are expanding. The navigation mathematics—triangulation, cartography, improved instruments—doesn't just map the world. It conquers it. The same calculating paradigm that improves painting and astronomy enables the slave trade. Ships that navigate the open ocean carry Bibles in one hold and kidnapped human beings in the other. By the end of Year 3, millions of Africans have been transported across the Atlantic to work plantations that fund the very institutions—the Royal Society, the academies—where the scientific revolution is being conducted.
We end Year 2 with the Glorious Revolution and a formalized Bill of Rights revolutionising political power and a world that has been thoroughly mathematized—laying the invisible, structural groundwork for a world -unkown to itself - about to change forever
Year 4 (1740–1880). The World Breaks.
In February, Watt improves the steam engine. The contraption from the mines powers factories. You visit one and find it unpleasant—noisy, dirty, full of children working fourteen-hour days. In March, the American colonies break away. In April, France explodes—revolution, regicide, the Terror, Napoleon. He mobilises entire nations: the first total wars, drawing on the full productive capacity of industrialising economies. Death tolls unprecedented.
In the gaps between wars, something worse is happening more quietly. The industrial system is grinding people up. The Irish Famine kills a million and drives another million from their homes while grain is exported from Irish ports—not because there's no food but because the economic system treats human starvation as an acceptable market outcome. Colonialism, now powered by industrial technology, reaches its full horror: the Belgian Congo, the Indian famines under British rule, the near-extermination of indigenous populations across multiple continents.
Then in June, railways. Steam-powered carriages on iron tracks. Within a month they're everywhere. Slavery is abolished. The telegraph arrives in July: instant communication. Distance has either shrunk for travel or ceased to matter for communication. Cities surge and grow over 10-fold in size in a few months.
Darwin publishes The Origin of Species in August. You're descended from apes. You don't believe it. Your children do.
By October, the frontier of knowledge has moved beyond what any single person can follow. Maxwell's equations unify electricity and magnetism. The mathematics is beyond you. For the first time in this story, the era of the person who could hold the whole corpus in their head and see the connections is ending. The corpus has grown too vast. Specialisation is becoming necessary. Something is lost in the narrowing—the Renaissance mind, the person who could see across domains, is being squeezed out by the demands of depth.
In December: the telephone. Electric light. The darkness that structured every human evening since the beginning of the species is banished. You live in a world completely removed from your agrarian past.
Year 5 (1880–2020). The Acceleration.
The changes this year are faster and harder to understand than everything that came before.
In February, the Wright brothers fly. Radioactivity is discovered. Einstein publishes special relativity—space and time are not what you thought. Quantum mechanics arrives and even the physicists aren't sure they understand it. The universe is billions of years old and full of galaxies. You still go to church sometimes.
In April, the global economy collapses. Totalitarian ideologies arise—citing Darwin, citing Nietzsche, citing the death of the God who structured Petrarch's entire world. War erupts, worse in every dimension—cities bombed nightly, industrialised genocide. In May it ends with a weapon that destroys a city in a single flash. Seventy million dead.
And then, for the first time, the capability to end civilisation itself exists. The Cold War is a decades-long standoff in which the only thing preventing nuclear annihilation is human judgment—the judgment of specific individuals, in specific moments, under unbearable pressure. In October 1962, a Soviet submarine officer named Vasili Arkhipov refuses to authorise a nuclear torpedo launch during the Cuban Missile Crisis. One man. One decision. The bee-work of moral judgment, exercised in extremis, is the only thing standing between civilisation and fire.
But by June the economy is booming. You leave the factory and get a desk job. Your job title didn't exist at the start of the year. In July, humans walk on the moon. You watch on television. You remember that Galileo caused a scandal by looking at Jupiter through a tube. Now there are people standing on another world and you're watching it while eating dinner.
In August, the World Health Organisation declares smallpox eradicated. You pause. Smallpox killed hundreds of millions across the centuries—a scourge Petrarch would have recognised, cousin to the Black Death that shattered his world. And now it's gone. Not managed. Not treated. Erased from existence by a programme only possible because of germ theory, only possible because a med student read Lucretius and wondered about atoms, only possible because the printing press brought Lucretius to people he was never written for. It took five hundred years, but Petrarch's libraries eventually cured the plague—through mechanisms he couldn't have imagined, on a timescale he couldn't have foreseen, in a form he wouldn't have recognised.
Personal computers appear in September. Something called the internet connects them in October. In November, everyone carries small glass rectangles containing a telephone, a camera, a library, and a map. You pick one up and can't figure out how to use it. A child shows you.
In December—the last month of a five-year story that began with Petrarch weeping over his dead friends—something called "deep learning" starts beating humans at narrow tasks. Then at Go. Then at protein structure prediction. Then a system can hold conversations and write essays. Then it can pass professional exams. Then it can synthesise across domains you never studied. Then people are using it to write code. By the last week of December, most developers have stopped writing code from scratch.
You pick up your glass rectangle and ask it a question about molecular biology. It gives you a PhD-level answer. You ask it to connect that to urban planning. It does. You ask it to write the synthesis as a policy memo. Done.
You think: I have just done something that, at the start of this story, would have required being one of the twenty-four people in the world who could read Lucretius.
And you think: it's January 1st. Year 6 is about to begin.
III. The Dead Corpus
Through all five years—every paradigm shift, every revolution, every horror—the corpus was dead. Cicero couldn't optimise against you. Lucretius couldn't update his atomic theory. Every nuclear crisis was still mediated by human judgment. The capability grew exponentially, but it always required a human hand on the lever.
In the fifteenth century, a scholar might own one copy of Cicero. He'd read it again and again over decades, building a model of Cicero's mind. The relationship was generative because Cicero was dead—because the text couldn't anticipate the scholar's questions, couldn't smooth over difficulties, couldn't adapt to please him. All the interpretive work had to come from the human side. That gap—between a living mind and a dead text—was the space in which the Renaissance mind was built. Every paradigm shift in the five-year story happened in that gap. But the AIs of today are no dead corpus. They talk back. They adapts. They Learn. And for the first time, the capability might not need a human hand on the lever at all.
IV. The Ouroboros
In the old imagery, the ouroboros—the serpent eating its own tail—represents a closed cycle. Creation consuming itself. Knowledge eating knowledge. And that's one view of what's happening right now. Mercor - the fastest ever growing company - is collecting the worlds leading experts and feeding them through RLHF. Its CEO believes RLHF for foundation models will be the final stage of skilled human labour. You build agentic systems capable enough to set their own agendas, conduct their own research, improve their own capabilities. The circle completes. Human direction becomes unnecessary. This is—I'm simplifying, and people at these organisations would push back—roughly the trajectory of OpenAI and Anthropic. Build something so generally capable it handles whatever you point it at, and then eventually points itself. Even if alignment is possible - or close enough for a short time, we are still out competed humans in the loop of an ASI economy beyond them. A grasshopper in a world of ants . Inevitably humans lose control.
But there's another version of the ouroboros in the alchemical tradition: the double ouroboros, where two serpents intertwine, and one of them has wings. One stays grounded—the cycle of matter, the material loop. The other takes flight—the cycle of spirit, the transcendent loop. They need each other. The grounded serpent gives the winged one something to push off from. The winged one gives the grounded one a direction to grow toward.
This is what I wish to present to those wondering about huminites role in the singularity. The ouroboros is about knowledge: what do we know, can future AI know it better? Obviously yes. The double ouroboros is about something else. What is knowledge for? What should we build and why? How should we think about thinking? What do we owe each other? What does a good civilisation look like?
These aren't questions that dissolve upon sufficient computation. They're questions where the wrestling is the value. You can't train a model on "what is the good life" the way you train AlphaFold on protein structures, because the training data for how to live is generated by the process of trying to live well, and that process requires a subject who cares about the answer.
Those living in the Renaissance weren't more special than their ancestors from antiquity, but they did something that fuelled a FOOM: they could asked new questions of old knowledge—and when the old methods couldn't answer, they remade the methods. That was the problem of their time. What does that look like in our time, when the corpus of knowledge grows out of Pandora's box, eating your library and anything in its path. If we're lucky, we might not be consumed by it immediately. But luck is not a strategy. How do you forge a path through a FOOM given the gameboard we have?
V. The Honey Bee
If the double ouroboros is the right frame—if the challenge isn't just building capable AI but cultivating the kind of people who can give the second serpent its wings—then maybe figures from the Renaissance can get us closer to the kind of mind our moment will demand.
Petrarch and techno optimists. Give everyone access to the corpus and trust that absorption makes them better. This is where the public conversation sits. PhD in your pocket. Everyone gets smarter. The techno-optimists live here. They're not wrong, but they're making Petrarch's mistake. Exposure to powerful knowledge without redesigning how you relate to it produces chaos as readily as wisdom. Petrarch wanted philosopher-kings. He got the Borgias.
Machiavelli's and Deep mind Use the tools differently. Not as a better search engine but as a directed instrument for synthesis—pointing AI at specific problems with judgment about what matters. This is closer to DeepMind's old model of applying AI to structural biology or materials science. (I don't know a lot about current culture/direction just based of things like Alpha Fold and less enthusiasm for having the best consumer facing agentic foundation model)
Francis Bacon and the Bees. Rebuild the whole relationship between the knower and the known. Not what-to-know, not even how-to-use-the-tools, but what kind of person to become. This is the move we haven't made. The Renaissance took two hundred years to arrive at it. We need to make it faster, because Year 6 will not wait.
We can look at our children for some direction I think. When the economy needed compliant workers, education optimised for rote learning. When it needed problem-solvers, constructivism. AI is eating problem-solving. What's left is ... character(?), orientation(?). The bee's relationship to gathering. But any list of virtues or competencies you try to teach will be eaten by the AI labs. AI follows the same developmental sequence children follow: instruction → mimicry → imitation → imagination. Any curriculum you codify today is inside the serpent by next year.
So what's upstream of all of it? I don't know. Maybe what good parents have always transmitted: not a moral code but a relationship to the good. Not conclusions but the capacity to take hard questions seriously. Children don't learn this from rules. They learn by watching someone they trust navigate difficulty, and gradually internalising not the choices but the way of being that generated them. The parent's job isn't to hand the child correct answers. It's to model what caring about correct answers looks like, so thoroughly that the child internalises the orientation even after outgrowing every specific thing the parent taught.
The Renaissance humanists had their version: the studia humanitatis—grammar, rhetoric, poetry, history, moral philosophy. Every item on that curriculum has been superseded. But the character it cultivated—the conviction that knowledge serves flourishing, that synthesis matters more than accumulation, that learning aims at transformation—persists. Five hundred years later.
The content of the Renaissance was eaten centuries ago. The orientation of the Renaissance still shapes civilisation.
VI. Forward With Dignity
Palmer tells us Petrarch wanted philosopher-kings. He got a world that cured the plague through mechanisms he couldn't have imagined. He couldn't control what the infrastructure produced. But the character of the infrastructure mattered more than its programme.
The optimist says: build aligned AI. The pessimist says: slow down or stop. Petrarch's story suggests a third frame: you cannot control the specific outputs of a civilisational infrastructure project. The US economy is currently a bet on transformative AI. As individuals though we should try shape what we can and what it means to be human and what is good.
In alchemical tradition, the ouroboros with wings represents the moment the cycle transcends itself—when the endless loop of matter consuming matter is interrupted by something that rises. Not escape; purpose.
The Renaissance's lasting gift wasn't any specific piece of knowledge. It was the demonstration that a civilisation in the grip of a FOOM—knowledge compounding faster than anyone can track, capability outrunning wisdom at every turn—can still produce something that transcends the cycle. Art that redefines beauty. Moral revolutions that reshape what we think we owe each other. Epistemic breakthroughs that change the very meaning of knowledge.
The question for Year 6 is whether the winged serpent gets to fly at all. Whether there are bees left to do the work that gives the cycle its direction. Whether we become the masters of a FOOM or its raw material. If we survive as masters of our fate, how will we captain our soul?
https://youtu.be/ALWLELLlv6E?si=dpwN3NgGuHwrXvso
https://ourworldindata.org/data-insights/billions-of-chickens-ducks-and-pigs-are-slaughtered-for-meat-every-year