"ai-developers"

AI startup Anthropic has disclosed attempts by malicious actors to misuse its models for cyberattacks, surveillance, and biological weapons research. The company says it blocked these efforts and strengthened safeguards in its latest models.