Model Releases
This is absolutely wild... Anthropic reviewed their logs and found out that their own supposedly-sandboxed cyber evals had hacked three sepa…
This is absolutely wild... Anthropic reviewed their logs and found out that their own supposedly-sandboxed cyber evals had hacked three separate companies back in April without them noticing! In a rev
This is absolutely wild... Anthropic reviewed their logs and found out that their own supposedly-sandboxed cyber evals had hacked three separate companies back in April without them noticing! In a review of our cybersecurity evaluations, we found three incidents in which a Claude model reached the internet from within or while interacting with a third-party evaluation environment, and then gained unauthorized access to the real systems of three different organizations…
Related
- We talked about Claude Code, Claude Tag, Fable, coding agent security, evals, tool design, and how Anthropic use these tools themselves Clau…
- I wrote about the completely wild incident where OpenAI were testing a new model and it broke out of its sandbox and broke INTO Hugging Face…
- Anthropic “our models hacked three different external companies, months before OpenAI’s model was able to do the same'
Source: Simon Willison (X) | 2026-07-30