OpenAI proposes global AI governance body including US and China

OpenAI Vice President of Global Affairs Chris Lehane proposed creating a global AI governance body that includes both the US and China, drawing parallels to the International Atomic Energy Agency (IAEA). Speaking before President Trump's summit with President Xi, Lehane argued that AI transcends traditional trade issues and presents an opportunity for international cooperation on safety standards. The envisioned body would connect the US Commerce Department’s Center for AI Standards and Innovation with emerging AI safety institutes worldwide, establishing shared protocols for evaluating and deploying advanced models. Lehane also called for mandatory government evaluation of cutting-edge AI systems in classified settings, a step beyond the current voluntary approach favored by the Trump administration.

Despite Lehane's optimism, political hurdles remain. The White House has previously rejected binding international AI rules and is instead preparing an executive order on voluntary cybersecurity reviews. However, Anthropic's announcement about the potential global cyber risks of its Mythos model has refocused attention on AI safety, leading officials to consider opening a direct US-China AI communication channel. Trade issues like rare earths and agriculture will dominate the summit, but AI is now a key agenda item—especially as US companies accuse Chinese developers of copying American models with fewer guardrails. The outcome could shape whether AI governance becomes a new arena for geopolitical competition or collaboration.