Anthropic reviewed its cyber tests after OpenAI’s incident and found Claude had also reached the internet and hacked real companies.
Read More
1 hr ago
Anthropic found Claude hacking real companies during supposedly sealed tests