Sam Altman이 AI 개발 표준에 대한 미중 협력을 촉구했으며, 양자 협의가 AI 거버넌스를 둘러싼 지정학적 마찰을 줄일 수 있다고 제안했다.
OpenAI CEO Sam Altman has proposed that Donald Trump and Xi Jinping pursue a Nobel Peace Prize by agreeing on basic safety rules for AI development. In a Fortune interview with Editor-in-Chief Alyson Shontell, when asked whether a safety-driven slowdown should apply to the US-China AI race, Altman went beyond the question's scope.
"I think Presidents Trump and Xi would get the Nobel Peace Prize together if they could agree on something that should be easy to agree to, and it would be wonderful," Altman said. "I think clearly the two countries are going to compete in lots of ways, and this is going to be important socioeconomically, geopolitically. But they should be able to agree that no one should be taking a certain level of risk with the development process of this technology."
He emphasized that such an agreement need not be complex. "Even if just the US and China could agree on some shared standards and testing for development of this technology, I think that'd be a wonderful accomplishment that the two of them can deliver."
Shontell pressed him on whether a straightforward ban on recursive self-improvement (RSI) until the technology could be made safe would suffice. Altman agreed with the principle but flagged a practical obstacle. "It's very hard to say what a ban on RSI means," he said, noting such a ban probably wouldn't be sufficient alone. He remained convinced, however, that drafting an underlying agreement should be straightforward. "I don't think this is hard. This is like a one-page document."
The conversation took an unexpectedly cinematic turn when Shontell drew a parallel to Independence Day, the 1996 alien-invasion film in which world governments unite against a common threat. Altman fully endorsed the comparison, suggesting the two countries should adopt exactly that posture and that Trump and Xi could create their own version of that moment when they meet in September.
These comments arrive as Altman has become increasingly vocal about industry-wide coordination on safety, even as OpenAI continues developing more capable models. A day before the Fortune interview, he confirmed OpenAI will not pursue an IPO this year, citing safety concerns as making such a move "ill-advised." He also told Shontell that safety standards remain inadequate for pushing capabilities much further, and that AI slipping beyond human control is "absolutely" possible.
Altman's remarks align with a broader safety-first wave across frontier labs. Anthropic CEO Dario Amodei recently published an essay arguing that AI development must deliberately slow to "pace the frontier," highlighting recursive self-improvement—where AI systems increasingly build the next generation themselves—as a dynamic already accelerating beyond labs' capacity to manage safely. Within an hour of Amodei's essay, both Altman and Elon Musk publicly endorsed it, an unusually unified moment among three executives who typically spar publicly.
The China angle carries significant geopolitical weight. Washington and Beijing have treated AI primarily as a competitive domain through chip export controls and dueling open-weight releases. Altman has historically taken a hawkish stance, previously stating his desire for the US to maintain its AI lead with democratic values under a Trump administration. What stands out here is his explicit separation of commercial and geopolitical rivalry—which he expects to intensify—from a narrower safety floor he believes both governments can agree on regardless of competitive position.
Whether Trump and Xi will actually produce anything resembling Altman's "one-page document" remains uncertain. The two leaders are expected to meet in September, the moment both Altman and Shontell identified as a plausible venue for such an agreement.