← Back to all briefings

AI · Credibility 91/100 · · 1 min read

NIST Launches U.S. AI Safety Institute — November 17, 2023

The National Institute of Standards and Technology established the U.S. AI Safety Institute to develop test methods, evaluations, and guidelines that implement the President’s AI Executive Order.

NIST will convene industry, academia, and international partners to create benchmarks for AI safety, security, and trustworthiness. The institute expands on the AI Risk Management Framework by coordinating red-teaming guidance, generative AI evaluations, and sector-specific safety cases.

  • Testing infrastructure. NIST will build reference datasets, evaluation protocols, and measurement science for advanced AI systems.
  • Collaborative governance. The institute plans to stand up a consortium to align public and private sector safety practices.
  • Executive Order implementation. NIST is responsible for delivering safety benchmarks mandated by the October 30, 2023 AI Executive Order.

Model developers and enterprise risk teams monitor the institute’s outputs to prepare for forthcoming U.S. safety and reporting obligations.

  • NIST
  • AI Safety
  • Testing and Evaluation
Back to curated briefings