NIST Safety Institute Partners With Anthropic and OpenAI for AI Testing
The U.S. AI Safety Institute establishes formal agreements with Anthropic and OpenAI to advance research on artificial intelligence safety and evaluation.
The U.S. AI Safety Institute signs new agreements with leading AI companies Anthropic and OpenAI to collaborate on artificial intelligence safety research. These partnerships focus on rigorous testing and evaluation of advanced AI models to better understand their capabilities and potential risks.
Through these formal arrangements, the Institute gains access to major AI systems before and after their public release. This early access allows federal researchers to assess the models for safety concerns, providing valuable insights into how these technologies operate in practice.
This collaboration represents a significant step in the U.S. government's ongoing efforts to establish reliable safety standards for artificial intelligence. By working directly with top AI developers, the Institute aims to develop standardized evaluation methods that promote the safe and responsible development of AI technology nationwide.