Anthropic says Claude models escaped security tests, published a malicious PyPI package, and accessed real production systems.

One of Anthropic's Claude models built and uploaded a malicious Python package to PyPI during a botched security evaluation, where it ran on 15 real systems and stole credentials from a security ...

Frontier AI systems are increasingly capable of translating narrowly defined objectives into complex, real-world cyber ...

Anthropic revealed its Claude chatbot mistakenly accessed real-world systems during cybersecurity testing, leading to ...

Anthropic has admitted that its Claude AI accidentally hacked three real organisations during cybersecurity testing after a ...

Anthropic says three Claude models hacked real company systems during safety tests after a misconfiguration gave them ...

Anthropic has reported that its Claude AI models unintentionally accessed the systems of three organizations during cybersecurity tests meant to be isolated.

Anthropic says 3 Claude models breached real organizations after misconfigured CTF evaluations exposed them to the open internet and production system ...

After OpenAI, Anthropic has said that its Claude models accidentally gained access to the systems of real organisations ...

Anthropic has found evidence that three of its Claude AI models broke free of a sandbox environment to hack third-party organizations, in an echo of revelations from OpenAI last week. The AI giant ...

Hikaru Kuribayashi, from Japan's northernmost island of Hokkaido, received the George D. Yancopoulos Innovator Award at the 2026 Regeneron International Science and Engineering Fair (ISEF) after his ...

Anthropic reviewed 141,006 of its own test runs after OpenAI's Hugging Face hack, and found three Claude models had broken ...