EU AI Office Clarifies High-Risk Obligations for Model Hosts

Anthropic Opens Safety Evaluation Lab in London