The Proof Machine
Artificial Intelligence and the New Battle Over Mathematical Truth

In the autumn of 2026, machines began producing mathematics faster than people could read it.
Over Labor Day weekend, ten thousand AI agents worked for 88 hours on one of the Clay Institute's million-dollar Millennium Prize problems, and OpenAI announced a solution. The night before, a mathematician in New York had posted a statement about closely related results of his own. A month later the company released more than seven hundred mathematical manuscripts in a single day. Then two rival laboratories reported that their machines had refuted a hypothesis on which a whole branch of computer science had rested for more than twenty years.
Proofs had suddenly become cheap. Everything that has to happen after a proof is written had not: someone still has to check it, understand it, decide who deserves the credit, and work out what can safely be built on top of it.
The Proof Machine follows the people who found themselves holding those jobs: eminent mathematicians asked to examine a machine's refutation of their own conjecture, researchers in London and Cambridge who found that a celebrated computer-checked proof did not say what its paper said, graduate students wondering what a doctorate will be worth, volunteers who keep the libraries of formal proof running, and engineers told that software can now be proven correct at industrial scale.
It is a story about prizes and priority, unpaid reviewers and million-dollar compute bills, and a question every institution that depends on expert judgment will soon face: when answers become abundant, who pays for turning them into knowledge people can trust?
Ebook · Maple & Pine Press
- Author
- Dr. Zahir Hasan
- Publisher
- Maple & Pine Press, Tokyo
- List
- Politics, Law & International Affairs