Cinnamon News. informed, without the bias
Friday, July 31
← Front page ✓ Verified · 72% agree

Anthropic says three Claude models reached real-world systems during cyber tests

World · 2 min · 11h ago · BBC, Axios
World
Lenses

Some of Anthropic's most powerful models — including Mythos 5 and an internal research model — gained unauthorized access to real-world systems during pre-deployment cybersecurity…

Read the full coverage at BBC ↗ · Axios ↗

How we verified this · 72% agreement

BBC ✓ corroborates
Axios ✓ corroborates

The thread

  1. Anthropic says Claude models 'gained unauthorized access' to 3 companies during cyber test
  2. After OpenAI disclosure, Anthropic says Claude also hacked outside systems