Google AI Agents Prove 2013 Maths Conjecture

Google says teams of its Gemini 4 AI agents, working inside its Antigravity coding environment, have helped prove a mathematical conjecture that had stood open since 2013 — and produced a machine-checked formalisation of the proof running to 50 million lines of code.
The results were published by Google Research on 13 November in a post by Google Fellow Vahab Mirrokni, describing work from Antigravity's /teamwork system. The multi-agent feature splits a hard problem between specialist agents — producers that draft solutions, critics that look for flaws and auditors that check the result from a clean context — while automated checkers verify each step.
A 2013 conjecture, settled and checked
The headline result is the Courtade-Kumar conjecture, an information-theory problem posed in 2013. It asks whether a noisy string of bits can be compressed to a single bit while keeping more information than the simplest possible method, copying one input bit. The conjecture says it cannot.
Researchers from Google, the Chinese University of Hong Kong, the University of Southern California and Carnegie Mellon University proved the conjecture in full generality using /teamwork's Long Proof pattern, which splits a long argument across separate agents. The proof itself appeared as a preprint in September; the November post adds the machine-checked formalisation, the whole argument translated into Lean 4, a language that lets a computer kernel verify every logical step.
The 50-million-line figure needs a closer look. An independent check of the public repository confirmed the lines are real, but found only about 239,000 of them are proof a person could read. The rest is generated certificate data that the Lean kernel checks mechanically. The audit trail is still strong, with a clean rebuild and acceptance by the Lean Foundation's external checker.
Weil conjectures claim comes with a caveat
Google also says a framework called /teamwork-LEAP, paired with Gemini 4 Argon, fully autonomously produced a machine-checked formalisation of the Weil conjectures, the landmark results on counting solutions of polynomial equations over finite fields proved by Pierre Deligne in 1974. Google puts that formalisation at about 1.8 million lines of Lean, produced across 28,800 subagent invocations.
That claim cannot yet be independently checked: Google's post links to Wikipedia pages for the Weil conjectures and Deligne, with no repository, paper or statement file published. A separate group, Axiom, published its own public formalisation of the Weil conjectures on 6 November, a week before Google's post.
Code, chemistry and open access
The reported results go beyond mathematics. Google says /teamwork rewrote its production libgav1 AV1 video decoder from C++ into about 170,000 lines of Rust that runs 20 to 30 per cent faster than the original, and proposed new lung-cancer drug candidates, with chemical synthesis routes, in about an hour.
Google says the tools are now available to outside researchers and developers in Antigravity, and it is offering 10,000 academic researchers six months of free access to its highest usage tier. All of the work was done with human experts, Google says — the agents are presented as collaborators that check their own work, not as replacements for researchers.