AIThe Recordabout 2 hours ago
Irregular silent on wider AI hacking impact
1 min read
Irregular, the cybersecurity firm behind AI hacking tests for Anthropic, OpenAI, and Meta, declined to say if other clients faced the same flaw. The company called the incidents "the exact same evaluation-environment issue" with "no current open issues." Meta confirmed Irregular's involvement this week.
Level
Hype check
Tap to vote and see what everyone thinks.
