Blog — Latest in Math Tech & AI

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

92 articles available Original analysis and reporting
Research & Studies 2 min read Nov 2, 2026

MFF-Agent tests an AI workflow across the foundations of mathematics

A new preprint combines language-model planning with Rocq verification to formalize selected tasks in set theory, real numbers, logic and the axiom of choice.

Read article 61
Research & Studies 2 min read Oct 30, 2026

AI search and Lean checks help build new quantum-code families

A new preprint combines LLM-guided program evolution, exact code checks and researcher-led proof work to develop entanglement-assisted quantum error-correcting codes.

Read article 152
Research & Studies 2 min read Oct 29, 2026

FARMS proposes a Lean-linked foundation for AI in computational mathematics

A new preprint connects numerical-linear-algebra and PDE statements to their sources, computational meaning, and Lean-checked implications for reusable AI workflows.

Read article 97
Research & Studies 2 min read Oct 29, 2026

Ramanujan Challenge report tests AI evaluation of research-level mathematics

A new preprint examines AI, computer-algebra, and Lean submissions to number-theory problems—and the difficult task of checking their proofs across formats.

Read article 83
Research & Studies 2 min read Oct 29, 2026

OpenConjecture tests AI on mathematics beyond headline breakthroughs

A new study samples 200 recent mathematical conjectures to examine not only whether frontier GPT models offer answers, but whether those answers match the intended mathematics.

Read article 93
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 128
Research & Studies 2 min read Oct 27, 2026

AI-assisted work proposes polynomial-time counting algorithms for spin glasses

A new preprint reports approximate-counting algorithms for several spin-glass models, with an unusually detailed account of model assistance and a companion Lean project.

Read article 100
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 58
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 211

Showing page 1 of 11 · 92 total