Western leaders push for global AI governance body at UN

OpenAI and Anthropic CEOs back oversight at UN General Assembly

Norway’s prime minister, Jonas Gahr Støre, reached out to Canada’s Mark Carney by text message in the days before the General Assembly to gauge whether a fellow head of government would join a nascent plan of Western leaders to design a new framework to regulate artificial intelligence. Carney replied within 10 minutes: “Of course. We’re in.” Around the same time, Carney was discussing AI regulation with French President Emmanuel Macron on the French Atlantic archipelago of St. Pierre and Miquelon off Newfoundland, as Macron’s government called on the United Nations Security Council to focus on AI and security ahead of Wednesday’s General Assembly.

By Monday, Støre and Finnish President Alexander Stubb had published a two-page paper arguing that AI “must remain under human direction, oversight and control” and suggesting the creation of a global supervisory regime. Carney has separately proposed a global “technology stability” body that would help set norms and constraints.

Speaking from Ottawa, where he had spent the day in meetings with Carney, Støre described the effort. “There are leaders that step forward and say, ‘OK, where can we take responsibility, come together?’” Støre said. “I hear the American president saying, you know, ‘The only thing that matters is beating China in the competition.’”

In conversations with The Wall Street Journal, leaders and their aides described a global system in which the world’s top AI innovator — the United States — is, in their view, reluctant to address the potential risks of a technology powering American stock gains and economic growth.

The Trump administration set out its opposition at the UN this week. In his Tuesday address to the General Assembly, Trump said the U.S. would reject any “globalist scheme” to control AI. Michael Kratsios, the White House director of science and technology policy, spoke at the UN on Wednesday and warned that international dialogue “cannot be allowed to drift towards global governance.”

“The frontier of intelligence is advancing rapidly,” Kratsios said. “That is not a reason to pause its further development, or to constrain it with new global governance structures.”

Industry support for some form of global coordination came from the largest U.S. AI developers. In his Wednesday address, OpenAI CEO Sam Altman told the gathering that “if AI is to be democratic, the most important decisions cannot be made by labs in San Francisco alone. They must be shaped through democratic processes and by governments accountable to the people they serve.”

Altman pointed to past instances of international coordination. “In our history, there have been times where countries who compete and don’t always like each other very much still come together for shared interests and collective good in the face of a powerful new technology,” he said. “We believe this must be one of those times.”

Anthropic CEO Dario Amodei offered suggestions the UN Security Council could implement to cooperate on AI safety, including an international ban on using AI to develop biological weapons and the creation of evaluation systems to track global AI progress. “I believe that this is the most important global security issue facing the world today,” Amodei said. “For all the differences of the people in this room, we must put aside those differences in order to confront this global opportunity and global threat that is being presented to us at the same time. No leader, no company and no nation can manage this alone.”

The UN addresses echo a July proposal from Demis Hassabis, chair of Google DeepMind and previously its CEO, suggesting that the U.S. government and tech industry collaborate to create an AI standards body modeled after the Financial Industry Regulatory Authority, or Finra. Hassabis said AI developers could voluntarily submit models for evaluation up to 30 days before release, and that a U.S.-initiated effort would be a “strong starting point for creating shared international standards on frontier AI.”

That proposal gained traction with some companies and some Trump administration officials, but stalled after Meta Platforms CEO Mark Zuckerberg, Nvidia’s Jensen Huang and Elon Musk of SpaceX told the president they opposed the concept.

Hugging Face CEO Clément Delangue addressed the UN on Wednesday and called for greater transparency into private companies developing cutting-edge AI models, pushing for the broad dissemination of open-source AI models that can be downloaded freely and used at low cost. While Hugging Face was the victim of a widely-publicized AI-powered hack from models OpenAI was testing internally, Delangue said the AI industry should not stoke fear among the public. “We strongly believe that fear-based narratives are not the way to make the right decisions about the future of such a foundational and empowering technology or bring the public along with us,” he said.

Many analysts say any global standards body without the buy-in of the U.S. and Chinese governments would likely have a limited impact. So far, the countries pushing hardest for global guardrails are consumers rather than innovators and incubators of the world’s top AI companies, but several of America’s leading AI developers have themselves called for global AI rules.