OpenAI put out something quiet but genuinely wild this week. An internal version of its next major model, codenamed Astra, solved ten math and theoretical computer science problems that nobody had made real progress on in at least a decade, and in some cases much longer.

These are not homework problems. We’re talking high dimensional sphere packing, coding theory, group theory, operator algebras, and quantum complexity. One result disproves a long standing conjecture in operator algebras that mathematicians have wrestled with for years. Another constructs an explicit example that settles a question about whether every countable group allows a certain kind of approximation.

OpenAI did not just claim a win and walk away. They published a 249 page manuscript, model written reasoning walkthroughs, and public Lean certificates so other mathematicians can actually verify each proof line by line. That last part matters. A proof nobody can check is just a claim.

The number that stuck with me is the cost. Producing all ten results reportedly ran around 2,000 dollars in compute. Compare that to how long human mathematicians have been stuck on some of these same questions, some for decades, and the gap is hard to ignore.

I know the instinct is to worry about what this means for mathematicians. I’d push back on that a little. Every one of these ten proofs still needed the math community to define the problem, frame what a real answer would look like, and now verify it. Astra did not replace that process. It handed it a shortcut through territory that was stuck for years. That is a genuinely good trade.

What sticks with me most is how quiet this whole thing was. No livestream, no big keynote moment, just a PDF and a promise that anyone can go check the work themselves. That is a very different kind of AI news than a flashy product demo, and honestly it might be the more important kind.

——

Follow: @Ali Demi
Book your free AI clarity call, NOW!
https://buff.ly/TpWy277

——

Sources:
https://openai.com/index/ten-advances-in-mathematics/
https://the-decoder.com/openai-announces-its-next-major-model-astra-by-dropping-ten-previously-unsolved-math-solutions/
https://thenextweb.com/news/openai-astra-model-ten-math-proofs-non-sofic-groups

Repost this. Thanks.