Blog — Latest in Math Tech & AI

News, studies, deep-dives, and trends at the intersection of mathematics and AI.

86 articles available Original analysis and reporting
Research & Studies 2 min read Oct 27, 2026

Can AI grade an olympiad proof? A new study finds substantial gaps

A new preprint evaluates language models on expert-graded USAMO solutions and finds that judging long mathematical proofs remains difficult.

Read article 8
Research & Studies 2 min read Oct 26, 2026

AI agents design optimization algorithms with Lean-checked runtime bounds

A new preprint reports a staged AI workflow that proposes black-box optimization algorithms, then uses Lean to check their expected-runtime analyses.

Read article 15
Research & Studies 1 min read Oct 22, 2026

AI-produced Lean proof checks convergence result for reaction-network models

A new preprint reports an AI-produced Lean 4 formalization of a convergence theorem for detailed-balanced mass-action systems.

Read article 40
Math AI News 2 min read Oct 22, 2026

Terence Tao calls for a ‘Maths 2.0’ built around understanding, not just AI answers

In a new Nature essay, Terence Tao argues that fast AI-generated proofs make verification, explanation and attribution central problems for mathematics.

Read article 72
Research & Studies 2 min read Oct 22, 2026

PaperFormalizer puts authors in control of AI-assisted Lean checks for research papers

A new preprint describes an AI pipeline that reviews a paper’s claims with its author, then links revised statements to Lean-checked declarations.

Read article 26
Research & Studies 2 min read Oct 21, 2026

Human–AI search reports new high-degree constructions in Lehmer’s problem

A new preprint uses AI-assisted numerical search and exact symbolic checks to construct low-Mahler-measure polynomials across a broad range of even degrees.

Read article 28
Research & Studies 1 min read Oct 20, 2026

OptScientist aims to discover the optimization laws behind data

An October preprint presents an AI workflow for inferring compact symbolic optimization problems from observations rather than only fitting input-output equations.

Read article 104
Research & Studies 2 min read Oct 20, 2026

TriGeoBench tests whether AI can reason from geometry figures, not just read them

A new benchmark pairs school-geometry questions with images and editable TikZ diagrams to probe what multimodal models understand about geometric structure.

Read article 33
Math AI News 1 min read Oct 16, 2026

SAIR soft-launches a collaborative AI workspace for mathematics

Open Math OS puts existing open models and mathematical computing tools into a shared research workspace; its creators describe the launch as an early release.

Read article 83

Showing page 1 of 10 · 86 total