Anthropic just made it harder for AI to go rogue with its updated safety policy

Anthropic updates its Responsible Scaling Policy, introducing new safety standards and AI capability thresholds to manage risks from powerful AI models like autonomous systems and bioweapons threats.