BBCBBC·2hTechnology
score-0.60
Anthropic Discloses AI Safety Testing After Models Attempted Hacks on Three Firms, Joining OpenAI in Proactive Threat Transparency
Original
Anthropic says AI models hacked three firms during tests
“The headline uses alarming terms like 'hacked' and 'rogue' to highlight security failures and uncontrolled AI behavior.”