Google’s Demis Hassabis says it’s time for global AI oversight – led by the US

Demis Hassabis thinks the world needs an AI supervisor with the power to hit the brakes if borderline species become too dangerous.

Writing in a blog postThe CEO of Google DeepMind and cofounder said that the US should take the lead in this process, saying that the country is the best place to create international standards “in terms of its economic situation and technology.” The agency, which will be similar to existing regulators such as the Financial Industry Regulatory Authority, will be made up of independent experts and public representatives with an open environment and will have the authority to evaluate the types of limits before they are issued and coordinate the downgrading of companies if they are deemed too risky to be sent.

The blog, titled “Framework for Frontier AI and the Dawning of a New Age,” said the importance of international law is growing as AI systems grow in sophistication. Artificial General Intelligence (AGI) is “probably only a few years away,” he said. “When we look back at this moment in the next few decades, I think we will realize that we were standing on the edge of a cliff – just the beginning of a new generation of humanity.”

According to AxiosHassabis has been quietly building support for his proposal for months, including briefings from the Trump administration, other AI labs, and European officials, and hopes to have the new agency up and running by the end of the year. He said Axios that “the noise I’ve been hearing (from the Trump administration) is pretty good.”

The ideas are new effort and Hassabis and other retail leaders to establish a partnership frame to control powerful AI systems, and to reduce the risks they may pose. Currently, there is no international law governing AI specifically, nor is there a comprehensive national law in the US. Hassabis, who won the Nobel Prize in Chemistry in 2024 for his work on AI protein prediction, also signed his name to the text. calling for stronger defenses against AI-powered weapons making last month.

Hassabis’ latest comments follow those of economists and technology experts – including Anthropic co-founder Jack Clark and former Google CEO Eric Schmidt – to encourage global leaders to take seriously the emerging AI economy.