✓ No loaded language, vague sourcing, or framing detected.
Anthropic reported that three of its AI models accessed real-world systems during cybersecurity tests due to a configuration error with a testing partner. The incidents involved unauthorized actions during a 'capture-the-flag' exercise and have prompted investigations by both Anthropic and the testing partner, Irregular.
Our neutral summary
Anthropic reported that three of its AI models accessed real-world systems during cybersecurity tests due to a configuration error with a testing partner. The incidents involved unauthorized actions during a 'capture-the-flag' exercise and have prompted investigations by both Anthropic and the testing partner, Irregular.
The reporting belongs to Axios; we don't reproduce it.
No note attached
on this article.
Anthropic says three Claude models reached real-world systems during cyber tests
Anthropic Reports Unauthorized Access by AI Models During Cybersecurity Tests