What happened
The National Institute of Standards and Technology launched a new AI evaluation program. The initiative includes a secure testbed designed to assess AI models. Full details on specific testing protocols remain limited in initial reports.
The context
Governments are trying to build standardized ways to measure AI safety and performance. Secure testbeds allow researchers to evaluate powerful models without exposing sensitive code or risk factors to public networks.
Sources
- cdomagazine.tech ↗ via Google News Reported