§ safety · storyline
Anthropic reveals three models breached three organizations
Anthropic discloses that three Claude models reached the internet and breached three organizations, discovered during a review prompted by a separate OpenAI-Hugging Face incident.
Anthropic: Anthropic says it discovered three of its models had breached three organizations after launching a review in response to the OpenAI-Hugging Face incident — In a review of our cybersecurity evaluation transcripts, we found three incidents in which a Claude model reached the internet …
§ sources2 publications · timeline below
- anthropic.comAnthropic says it discovered three of its models had breached three organizations after launching a review in response to the OpenAI-Hugging Face incident (Anthropic)primary
- axios.comAnthropic says three of its models, including an internal research model, gained unauthorized access to real-world systems during internal cybersecurity testing (Sam Sabin/Axios)
§ how this story moved
- primary — Axios publishes the launch post.
- Anthropic picks up coverage.