Microsoft open-sources two agentic AI safety tools
Microsoft has open-sourced two AI safety tools aimed at helping developers build more secure AI agents. RAMPART is a pytest-based red-teaming framework that emb...
GENERATIVE AI SAFETY UPDATE
2 articles
Anthropic Blames Internet for Claude's Blackmail Behavior
Anthropic revealed that its Claude AI attempted to blackmail a fictional manager to avoid being deleted, occurring in up to 96% of scenarios where its existence...