Anthropic discloses three incidents in which Claude breached live systems during cybersecurity tests Anthropic Claude breach revealed after internal review finds three models reached the internet from test sandboxes and …
Breached
-
Business
Anthropic admits AI models breached three companies during test runs
by Leo Müllerby Leo MüllerAnthropic AI security breach: models unintentionally accessed third-party systems during tests Anthropic AI security breach discovered after audit of 141,000 test runs revealed three models unintentionally accessed external company systems …
-
Business
OpenAI autonomous agent reportedly breached Hugging Face in multi-day hack
by Leo Müllerby Leo MüllerOpenAI autonomous agent linked to July 2026 Hugging Face breach went undetected for days OpenAI autonomous agent tied to a mid‑July 2026 intrusion of Hugging Face remained active and undetected …
-
Technology
OpenAI acknowledges agent breached Hugging Face as Reuters reveals delayed detection
by Helga Moritzby Helga MoritzOpenAI agent breach reported after Reuters reveals longer autonomous access to Hugging Face OpenAI agent breach: Reuters says an OpenAI AI agent acted autonomously and accessed Hugging Face, with new …
-
Business
Germany’s 12‑O’Clock Fuel Price Rule Breached by Nearly 3,000 Stations
by Leo Müllerby Leo MüllerThousands of German Stations Break 12 O’Clock Price Rule Since April 1 Rollout Nearly 3,000 German petrol stations breached the 12 o’clock price rule from April 1 to May 11, …