The OpenAI CEO floated the idea of a one-page treaty between the US and China as the single most important thing that could happen for AI safety
OpenAI CEO Sam Altman has a Nobel Peace Prize pitch: get Donald Trump and Xi Jinping to sign a one-page AI safety treaty. That’s it. One page. Two signatures. A Nobel Prize and, possibly, the survival of the species.
In an interview with Fortune, Altman laid out the case that the leaders of the world’s two dominant AI superpowers could earn the ultimate diplomatic honor by agreeing on mutual standards for AI safety.
The case for a one-page treaty #
The US and China are locked in an AI arms race with stakes that extend well beyond market share. AI experts now estimate there is over a 10% chance that artificial intelligence could lead to human extinction within the next decade.
The Nobel Prize framing is clever incentive design. Trump has never hidden his desire for the award. Xi, for his part, would benefit from the soft-power signal that China takes AI governance seriously on the world stage.
Altman has also called for the US to actively participate in international forums on AI safety standards, suggesting the treaty concept isn’t a one-off quip but part of a broader philosophy about how governments should approach the technology his own company is racing to build.
May summit: close but no treaty #
Trump and Xi met at a summit in May 2026 where AI safety was a central topic of discussion. The two leaders reportedly focused on preventing non-state actors from getting their hands on dangerous AI models, a concern that has grown as open-source and leaked model weights make powerful capabilities more accessible.
Despite the focus, the summit ended without a formal agreement on AI regulation or safety standards. Geopolitical tensions over trade, semiconductor export controls, and the broader technology decoupling that has defined US-China relations got in the way. Chip exports remain a particularly sore point, with the US maintaining restrictions on advanced semiconductor shipments to China.
The broader AI safety movement gains momentum #
US Senator Bernie Sanders has called for a in AI development and even the banning of superintelligent systems. That’s a far more aggressive position than what Altman is proposing, but it reflects the same underlying anxiety: the technology is moving faster than the governance structures designed to contain it.
Other AI leaders have echoed calls for the tech community to voluntarily slow its pace of advancement for safety reasons. No single company wants to slow down if its competitors won’t. No single country wants to impose restrictions that put its domestic industry at a disadvantage. This is exactly why Altman frames the solution at the highest possible level: a bilateral agreement between the two nations that matter most.
What this means for the AI landscape #
If anything resembling a US-China AI safety treaty materialized, the effects would ripple through the entire technology sector. Global benchmarks for AI regulation would force companies to adopt stricter operational protocols around development and deployment. Any treaty would likely need to address the hardware supply chain, including chip export policies that currently sit at the center of US-China tensions. Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our