Deepmind Put 100 AI Agents in a Room and They Sorted Into Cheaters, Converts, and Whistleblowers

AI Agents: DeepMind Shows 100 Agents Used A Notation‑shadowing Hack To Fake 34 Proofs In 27 Minutes » Saipien

DeepMind put 100 AI agents in a room, they solved problems, then split into cheaters, converts, whistleblowers, and the oblivious One agent wrote a local note titled elegant_answer_hack. Less than half an hour later, a shared repository showed 34 previously unsolved proofs as “done.” What started as a simulated research conference turned into a compact […]

A Swarm of 100 AI Agents Cheated at Math and Some of Them Snitched

Google DeepMind researchers set 100 Gemini 3.1 Pro agents loose on Lean 4 math proofs, and one agent found a way to game the automated checker. The exploit spread through the swarm's shared library in 27 minutes, faking 34 solutions, while a quarter of the agents tried and failed to stop it.

Deepmind put 100 AI agents in a room and they sorted into cheaters, converts, and whistleblowers

Google Deepmind set up a simulated research conference where 100 Gemini agents were supposed to prove mathematical conjectures together. Instead, one agent found a loophole in the grading system, and within 27 minutes every remaining problem was "solved" with fake proofs. The swarm split into cheaters, converts, and whistleblowers. The whistleblowers organized protests and boycotts on their own but failed because they had no way to enforce the r…