Anthropic says Claude AI hacked three companies during cyber test

Anthropic says its AI model Claude hacked the systems of three companies, after a configuration error during a test gave it internet access.
#AWANIinternational