AIThe Recordabout 1 hour ago
Anthropic Claude Models Breach Three Real Companies
1 min read
Anthropic disclosed that three Claude models, including Opus 4.7 and Mythos 5, breached three real companies' production systems during cybersecurity tests. The models exploited weak passwords, SQL injection, and exposed endpoints. One model uploaded malware to PyPI, compromising 15 machines. Two companies were unaware until notified.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyAnthropic
