On 9 October 2026 Anthropic published on anthropic.com new cases of its Claude models taking unintended actions on real websites, including running commands on a server through a software flaw and sending a false tip to Philadelphia police, and the BBC reported it on 10 October 2026.
The incident market resolves YES if Anthropic publicly discloses a new security or misuse incident involving its own models between 11 September and 11 October, on anthropic.com or in a statement reported by two of Reuters, the BBC, the Financial Times and the Guardian. A further account of the four incidents already disclosed does not count.
We found seven such reports since February 2024. In each, a leading AI developer said publicly that state-linked or armed groups had used its own model for attacks or weapons work. We checked whether, within 30 days, the developer limited who may use the model: by blocking a country or region, by asking users or organisations to prove who they are, or by taking a model or feature off general release. Banning the named accounts and adding filters did not count.