What happened
The National Institute of Standards and Technology released a new evaluation platform for artificial intelligence. The system helps developers test software performance and safety risks. It provides standardized tools to measure how models handle different tasks.
The context
Governments and companies lack universal benchmarks for testing AI systems. NIST sets official measurement standards for the US government. A unified testing platform helps build consistent rules for evaluating model risks.
Sources
- Nextgov/FCW ↗ via Google News Reported