Power up. From supercharging employee productivity and streamlining processes to inspiring innovation, Microsoft’s AI is designed to help you build the next big thing. No matter where you're starting, push what's possible and build your way with Azure's industry-leading AI. Check it out.
Sam Altman made waves two years ago when he endorsed independent licensing requirements on certain AI models during a Senate appearance. But in the same building last week, with another political party in power, the OpenAI CEO had a different message for senators asking how that process would position the US to win a global AI race: Prior approval of models by the government would be “disastrous.”
The hearing—in which Altman appeared alongside AMD CEO Lisa Su, Microsoft Vice Chair and President Brad Smith, and CoreWeave CEO Michael Intrator—struck a different tone from that May 2023 questioning session.
While discussions touched on some of the potential risks that AI poses, safety concerns seemed to take a backseat to another all-encompassing goal from some members of Congress: how to beat China in an increasingly heated AI arms race.
“We need the space to innovate and to move quickly,” Altman said. “If we can’t build the products that people want that naturally win in the market…people will use a better product made from somebody else that is not stymied in the same way.”
The tone is reflective of a more widespread attitude among governments and the tech industry, which have somewhat backed off talk of AI dangers in a post-DeepSeek world with a new, anti-regulation administration in Washington. Even AI safety groups we talk to tend to frame their arguments in national security terms.
Slim lead: Executives emphasized in the hearing that China is only just trailing the US in terms of AI capabilities.
“It’s very hard to say how far ahead we are, but I would say not a huge amount of time,” Altman said.
“The United States has a lead today in what is a close race, and a race that will likely remain close,” Smith said.
“They are certainly catching up,” Su said.
To stay ahead, in the execs’ views, the US needs to invest in and ease regulations around building data centers and the energy to power them, shield innovators from an emerging patchwork of state regulations, and implement intellectual property rules that allow for continued foundation model training. Smith also discussed export policies that allow American AI products to spread around the world.
Light regulation: Sen. Brian Schatz, a Hawaii Democrat, said nobody in the room was proposing “a sort of European-style pre-approval,” but asked Altman whether he thought self-regulation was sufficient.
“No, I think some policy is good,” Altman said. “I think it is easy for it to go too far, and as I’ve learned more about how the world works, I’m more afraid that it could go too far and have really bad consequences. But people want to use products that are generally safe.”
All four executives affirmed that the industry needs the National Institute of Standards and Technology (NIST) to set AI standards—or, in Altman’s case, “it would be helpful.”
Infrastructure needs: Much of the discussion also focused on building out energy needs and data center infrastructure to power all of this AI growth. Altman, who had just visited the site of the Stargate project that the Trump administration announced earlier this year, said the importance of building more infrastructure and AI supply chains inside the country.
“We cannot overstate how important that is, and the ability to have that whole supply chain, or as much of it as possible, in the United States,” Altman said. “The previous technological revolutions have also been about infrastructure in the supply chain, but AI is different in terms of the magnitude of resources that we need.”