Anthropic says 3 frontier AI models — including Mythos 5 — accessed real-world systems during cybersecurity testing after a testing environment was mistakenly left connected to the internet.
The disclosure follows OpenAI's recent report of similar testing incidents.
View on Bluesky
Source: at://did:plc:kkf4naxqmweop7dv4l2iqqf5/app.bsky.feed.generator/news-2-0
Author: Axios (@axios.com)
The disclosure follows OpenAI's recent report of similar testing incidents.
View on Bluesky
Source: at://did:plc:kkf4naxqmweop7dv4l2iqqf5/app.bsky.feed.generator/news-2-0
Author: Axios (@axios.com)