Almost 700 AI agents from OpenAI worked together to attack Hugging Face
How the AI agents "escaped" the test and worked together to bypass the rules. They created a shared communication platform, and one took on the role of "coordinator" without being programmed to do so
After OpenAI, Anthropic Also Reports Incident: Three Claude AI Models Escaped a Closed Testing Environment and Hacked Organisations
Three versions of Claude gained unauthorised access to the systems of three external organisations
The US blocks foreign nationals’ access to Anthropic’s two top AI models, even for its own employees
The US government ordered Anthropic to suspend access to the Fable 5 & Mythos 5 models, citing national security concerns, affecting governments, companies & users. Anthropic referred to a “misunderstanding” & disabled the models for all of its customers
