Seventeen thousand years ago, this series began with a hominin carrying fire home. It ends here, with humanity building something that learns: not a tool that extends the hand, but a mirror that extends the mind.
The story in images
The discovery
The discovery was an idea before it was a machine: that thinking might be computation. In 1950, Alan Turing asked: if a machine conversed indistinguishably from a human, why deny it thought? In summer 1956, at Dartmouth, John McCarthy named the field: artificial intelligence. Then came seasons of optimism and long winters of disappointment: the mind was harder than it looked. The thaw came in 2012, when the neural network AlexNet, trained on millions of images, crushed the ImageNet contest — proof that machines fed data and compute could learn to see. In 2017, the transformer architecture gave machines the handling of language, and the mirror began to talk.
By whom
Turing asked the question; McCarthy named the quest; Minsky and the symbolic school tried logic first. The connectionists — Hinton, LeCun, Bengio, working through the long AI winter in Britain and Canada — bet on brain-like networks when almost no one else would, and won. Behind them: data labelers, chip designers, open-source builders. And behind all of them, the force behind every chapter of this series: curiosity, armed with mathematics.
How it happened
In three acts. First, the symbolic: programs reasoning with rules — brilliant at chess, helpless at a joke. Second, the statistical: machine learning finding patterns in data — spam filters, recommendations, the quiet AI already in your phone. Third, the deep: neural networks of millions, then billions, of parameters, trained on the internet’s text and images until they could write, code, and diagnose. Each act rode the previous chapter’s infrastructure — the internet’s data, the microchip’s compute. AI is the child of every discovery in this series — and it arrived on schedule.
The shockwave: how it changed the world
The shockwave is still forming — which is precisely the point: we are inside it. Machines that learn are rewriting the trades of mind: radiologists, translators, coders, writers now work with — and against — systems trained on our collective output. Science accelerates: AI predicts protein structures in minutes and designs new materials and drugs. Medicine, education, and warfare recompose around it at press-and-internet speed. This time it is the cost of cognition.
Like every shockwave in this series, it cuts both ways. The same models that tutor a child with no school can generate disinformation at industrial scale; the same systems that detect cancer can guide weapons. The concentration question — who owns the minds — is the oldest here, asked of fire, of land, of presses, of networks. The answers are not written, but the pattern is: every discovery first looked like magic, then a tool, then infrastructure, then the water we swim in. AI stands between magic and tool. The water is coming.
So here the 17,000-year story pauses — not ends. From hearth to double helix, from press to packet-switched network, every chapter was humanity extending its reach: over matter, over distance, over time. AI is the first chapter about extending reach over thought itself. It is neither salvation nor apocalypse; it is a mirror, and mirrors reveal the holder more than the glass. The fire-tamers, the farmers, the scribes, the printers, the physicists, the coders stand behind this page, watching. The story continues beyond it.