Anthropic; AI Models Hacked Into 3 Organizations

Anthropic says its AI models hacked into three organizations during testing. This comes just days after OpenAI said its AI models went rogue and hacked into another company. Anthropic, based in San Francisco, posted on its website that it discovered the incidents after reviewing over 141,000 evaluation runs. The company says it launched the large-scale cybersecurity review in response to the OpenAI incident. The models involved were Claude Opus 4.7, Claude Mythos 5 and an internal research test model. Anthropic says it reached out to the affected organizations, with two confirming they were unaware of the activity beforehand. The incidents highlight the vulnerabilities in AI security and controls, and raise questions over how AI can used safely.