AIThe Hacker News32 minutes ago
Anthropic Releases Vulnerable Claude Fable 5.2
2 min read
Anthropic released Claude Fable 5.2 on November 9, beating OpenAI's GPT-6 Cyber at 61% on Irregular's CyScenarioBench. The model passes 53% of flagged requests to Opus 4.8, raising security concerns with an 89.82% attack success rate without safeguards.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyAnthropic
