AI Safety Assurance Is Still Goodwill, Not Evidence
The UN's first scientific assessment of AI says safety assurance still runs on developer goodwill, not verification. Here is what evidence-based governance would actually require of agentic systems.









