Axiom Math, which has recruited top talent from Meta, has raised $64 million in seed funding to build an AI math whiz. bit.ly/48DoSri?utm_ca…

Hear our founder Carina Hong on the latest @TWIML podcast with @SamCharrington 🎙️on why Axiom is bullish on math 📐🚀 “Math and coding are the two biggest parts of the digital world. Coding is heavily invested in. Math is not. Not because math isn’t trackable, but because math

Excited to see François Charton represent Axiom at the Kavli Institute for Theoretical Physics! Francois' work pioneered the three prevailing approaches in AI for math discovery: from generative and translative methods to world models. In this talk, he unveiled some recent

This year’s IA40 theme is the Reasoning Revolution. At Axiom, we’re building the reasoning engine. RL + verification = huge performance boost 📈 in formal mathematics. Our founder Carina Hong shared why there is no better moment to tackle this frontier.

Come listen to our CTO Shubho Sengupta talk about his journey in this latest podcast! From writing the first standard GPU libraries for CUDA to training large reinforcement learning models like OpenGo (the open-sourced version of AlphaGo), Shubho has been pushing every frontier

Excited to partner with Toyota Ventures and Chris Abshire !

You don’t need formal reasoning to write a poem. But when billions of dollars, personal data, or human lives are at stake? You do. Generative AI can guess. But in high-stakes systems, we need AI that can prove it’s right. That’s what Madrona Partner Jon Turow explored with

Incredibly impressed by and proud to be supporting Carina Hong and the Axiom team. François Charton puts it well: “One fields medalist called this either amusing or game changing. It wasn’t amusing, that was the beginning of AI for math!” Here’s to new discoveries!

If you care about the next frontier of reasoning — not just faster LLMs, but systems that formalize, prove, and conjecture — Axiom is building it. An invitation to work on the hard, beautiful problems. A key piece is construction-driven discovery: the examples that seed

The two most exciting approaches to AI for math, the competition has begun: Axiom: PatternBoost, Int2Int, Mathematical World Models. DeepMind: FunSearch, AlphaEvolve.

Very proud of my collaborator François Charton who is a trailblazer of AI for Math (and now also Theoretical Physics). Particularly pleased that our work arxiv.org/abs/2507.03732 was mentioned!

Congrats François Charton! This will likely be an amazing team! Lucky those who will be hired there

Had a great time hosting Carina Hong from Axiom on Built This Week. Axiom, who secured over $60M in funding earlier this month, is building a self-improving AI mathematician! One of the most exciting frontiers in reasoning and formal math. #AxiomMath #Axiom #AI #Math

Beautiful work from François François Charton at Axiom. Transformers don’t just approximate Collatz; they learn the hidden structure behind it. Models discover the control logic of long Collatz steps, a complex arithmetic function, and fail only in predictable, non-hallucinatory