Historic first as companies spanning North America, Asia, Europe and Middle East agree safety commitments on development of AI

New commitments to develop AI safely have been agreed with 16 AI tech companies spanning the globe, including companies from the US, China and the Middle East, marking a world-first on the opening day of the AI Seoul Summit on May 21. Where they have not done so already, AI tech companies will each publish safety frameworks on how they will measure risks of their frontier AI models, such as examining the risk of misuse of technology by bad actors. Companies will take input from trusted actors including home governments as appropriate, before being released ahead of the AI Action Summit in France in early 2025. The 16 companies who have agreed to these commitments are:

1. Amazon 
2. Anthropic 
3. Cohere 
4. Google / Google DeepMind 
5. G42 
6. IBM 
7. Inflection AI 
8. Meta 
9. Microsoft 
10. Mistral AI 
11. Naver 
12. OpenAI 
13. Samsung Electronics 
14. Technology Innovation Institute 
15. xAI 
16. Zhipu.ai