Anthropic's Claude formalizes Fermat's Last Theorem proof in 11 days
Anthropic announced that its AI model, Claude, has successfully formalized the proof of Fermat’s Last Theorem into computer-verified code. This achievement marks a significant milestone in mathematical AI, as the model translated the complex 1994 proof by Andrew Wiles and Richard Taylor into 13 million lines of Lean code. The process took only 11 days, a task estimated to require approximately…
Key points
- Claude formalized Fermat’s Last Theorem into 13 million lines of verified Lean code in 11 days.
- The task was estimated to take humans 10 years, highlighting a massive efficiency gain for AI.
- Experts view this as a major step toward AI verifying the entire library of mathematical knowledge.
Mathematicians, including Alex Kontorovich from Rutgers University and Kevin Buzzard from Imperial College London, described the result as a breakthrough that demonstrates the rapidly improving capability of AI to handle high-level mathematical reasoning. Buzzard noted that this task was an order of magnitude more difficult than previous AI formalization successes, such as the certification of Maryna Viazovska’s sphere-packing work earlier this year.
The development suggests that AI will play a crucial role in verifying existing mathematical knowledge and potentially discovering errors in established results. Experts believe that at the current pace of progress, AI systems may soon be able to scrutinize the entire library of mathematical proofs, fundamentally changing how mathematical validity is established and checked.
Coverage and discussion
1source- Hacker News discussion · 6 pointsnews.ycombinator.com
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- Nvidia pledges $250M in compute for UK AI agent safety research · 8 src
- OpenAI offers U.S. university labs free ChatGPT Business seats with Rosalind access · 2 src
- Anthropic reports small AI models may self-replicate on IoT devices within months · 29 src
- Researchers propose LLM-referenced C-test for evaluating supra-intelligence · 8 src
- Researchers study permutation diffusion LLMs for recommendation reranking · 1 src
Comments
via GitHub Discussions