Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Story

1,000 Claude Fable 5.2 agents reproduce a formal Fermat proof

AI summary

Anthropic reports that 1,000 agents running an early version of Claude Fable 5.2 reproduced a Lean-checked formalization of Fermat’s Last Theorem without Prove2Me or researcher steering. The team took 28.7 hours, used one-third as many output tokens, and produced a proof one-seventh the length of the earlier formalization. The toy experiment operated in an isolated sandbox with command screening and offline monitoring, so its results do not establish performance outside that controlled setup.

In-depth insight

The experiment provides a vendor-reported example of large agent teams organizing a long-horizon proof task with general-purpose tools, while also showing how sandboxing and layered monitoring enabled the run.

Safely scaling multi-agent teams with Claude Fable 5.2 / claude.dev Blog

Agents

Safely scaling multi-agent teams with Claude Fable 5.2

How a team of 1,000 agents running an early version of Claude Fable 5.2 reproduced a formal proof of Fermat's Last Theorem with general tools, a shared database, and layered safety controls.

AUTHOR

Barry Zhang, Tianyi Peng, Elle Najt, Keller Scholl, and Lance Martin

#Anthropic#Claude Fable 5.2#Prove2Me#Lean#Fermat’s Last Theorem

Original · 15 min readRead the original
Group chat

No comments yet. Start the conversation.