Anthropic Grants METR Researchers Unprecedented Internal Access to Evaluate AI Safety
Anthropic has provided METR researchers with employee-level access to evaluate its frontier AI models and internal practices. Under the agreement, METR will publish independent risk assessments without Anthropic retaining the right to censor the findings. This move marks a significant shift toward greater transparency and external oversight in the highly competitive artificial intelligence industry.
Source: TechCrunch