WASHINGTON POST: Donald Trump resists AI slowdown amid China fears and despite warnings from tech leaders

US President Donald Trump on Sunday appeared to dismiss calls from top artificial intelligence leaders to slow the development of AI models, as Democrats seize on his administration’s resistance to regulation as a line of attack in the midterms.

“We’re leading China in AI, we’re the most sophisticated country in the world, and frankly I want to keep it that way,” he said on the sidelines of the Irish Open golf tournament. “Whoever wins AI wins.”

Trump has staked much of his economic record on speeding up AI advances, as a boom in the technology drives historic highs in the stock market. But that hands-off approach is facing new scrutiny after three of the AI industry’s most prominent leaders — Anthropic CEO Dario Amodei, OpenAI CEO Sam Altman and xAI founder Elon Musk — endorsed slowing the pace at which the industry improves AI model capabilities.

Their united front followed the resignation of an Anthropic researcher who on Tuesday warned that the firm and its competitors were not responsibly developing AI models. The researcher’s statement inflamed public debate about the risks of the technology.

On Sunday afternoon (US time), Microsoft CEO Satya Nadella said his company welcomed ideas like deliberate pacing and embedding evaluators, with some caveats.

In a statement posted on X, Nadella emphasized that any safety regimen will also have to factor in enterprise customers’ need for privacy and flexibility. “(F) or firms, it’s imperative that they retain full control over their unique and tacit knowledge,” he wrote.

Trump’s remarks placed him once again firmly on the full-speed-ahead side of the debate over the pace of AI development, which has increasingly divided both the technology and political worlds. He faces growing isolation in his position, as even members of his own party and some allies in the tech industry have grown uneasy with Trump’s apparent lack of concern over potential AI dangers.

Trump attributed the warnings to unspecified “negative forces.”

“I think you have a lot of negative forces that are bringing it up that shouldn’t be bringing it up,” he told reporters. “And they’re bringing up things that won’t happen.”

Many hours later, on the flight from Ireland back to Washington, Trump insisted he was not downplaying the tech executives’ warning and said he had spoken with some of them.

“I’m not downplaying. But it’s going to be more good than bad — by a lot,” he said, referring to AI. He also said that he has used AI but declined to say how.

Democrats are seeking to capitalise on growing public unease about AI, with multiple polls showing Americans increasingly wary of the technology and its rapid development. The issue potentially ties together several Democratic lines of attack, allowing them to portray Trump as inattentive to his work and too close to billionaire leaders of the tech industry.

In a flurry of social media posts over the weekend, Democrats spanning the party’s ideological spectrum pledged to rein in the industry and accused the administration of failing to confront the technology’s mounting risks.

Abdul El-Sayed, a progressive Democratic Senate candidate in Michigan, warned Sunday that technological development was outpacing safety practices and endorsed a proposal from independent Senator Bernie Sanders to pause AI development.

“There are things that we should not be using AI to do, and we collectively have to decide that we need public oversight so that all of this technology is not being created for the interest of a few against the well-being of the many,” El-Sayed said in an interview on Face the Nation.

Josh Shapiro, the Pennsylvania governor, a centrist and potential 2028 Democratic presidential candidate, called the industry leaders’ warnings a “bright flashing red light.”

“It’s long past time for our Federal Government to step up, take this seriously, and put up some guardrails — both here at home and on the global stage,” Shapiro wrote on X, drawing parallels to the regulation of pharmaceuticals and airplanes.

Shapiro recently has grown critical of the industry after previously fostering construction of the data centres underpinning AI development. Last month, he signed an executive order restricting their development.

Former president Barack Obama last week also privately urged Democratic Party leaders to make AI regulation central to the party’s agenda. He encouraged them to focus on safety, the technology’s impact on children and its economic impacts.

Obama viewed the industry leaders’ calls for slowing down AI development as a “positive development,” according to a person familiar with his thinking, who spoke on the condition of anonymity to describe private conversations. The New York Times first reported Obama’s recommendation.

The Democratic attacks have posed a challenge for Republicans, who are increasingly aware of the voter backlash to AI but are running on a party agenda that is inextricably tied to the White House.

Trump has moved to clear regulations that his donors in the tech industry warned could slow AI development down. During his first week in office, he rescinded a 2023 Biden executive order that created a process for evaluating the risks of cutting-edge AI models.

As concerns about AI risks have grown this summer, the administration has developed a new framework for reviewing AI models for cybersecurity risks.

Some Republicans facing tight midterm races have expressed greater openness to AI regulation. Mike Rogers, a former Republican lawmaker who is battling El-Sayed for the Senate seat in Michigan, wrote on X on Sunday that he supports slowing down AI development and would prioritize passing a national security framework addressing the technology.

“My experience as Intel Chairman taught me the time to stop a crisis is yesterday,” said Rogers, who served as chairman of the House Intelligence Committee from 2011 to 2015.

Republican leaders have sought to place the onus on companies to protect the public, doubling down on the White House’s position.

“There is an obvious corporate responsibility that the people who are creating these models have to have to ensure that their products are safe,” Republican House Speaker Mike Johnson said Sunday on CNN.

Johnson said he would meet with the leaders of the tech platforms as a first step. He also pointed to a post from David Sacks, the architect of the Trump administration’s hands-off approach to AI regulation. Tech leaders should “go ahead” with pacing AI development, Sacks said, but he rejected the companies’ claims that the industry needed intervention from Washington to do that.

“Stop pretending you need anyone else’s permission,” Sacks wrote on X.

He also said that the companies should stop saying their interest in slowing down “is purely altruistic.” The companies would be liable if their products caused a significant cyberattack, he said.

Amodei and other company officials have asked the US Government to waive antitrust laws so that they can coordinate on matters of AI safety, and Amodei wrote in his essay that it would be helpful if the government would “mediate or at least enable these discussions.”

Some policy experts and tech leaders dismissed the call for an antitrust exemption, arguing the companies were advocating policies that were favorable to their financial interests but would not further the goal of protecting public safety.

Slade Bond, a former Biden Justice Department official, said courts for a century have allowed companies to cooperate for the purpose of public safety. Companies can’t invoke safety as pretext for suppressing competition for other, illegal reasons.

But a deal between AI companies to prevent a cataclysmal event, such as a financial crisis induced by a bot swarm, would be viewed as pro-competitive under the Sherman Antitrust Act, an 1890 statute that undergirds US competition policy.

Asking the government for a waiver suggests the companies are extremely risk-averse when it comes to potential liability, despite having massive market caps, legal teams and litigation budgets, said Bond, who was a lead counsel on the House antitrust investigation into the tech industry in 2019 and 2020.

“They want a blank cheque,” he said.

“They don’t like product liability, they don’t like copyright, and now they don’t like antitrust.”

Industry leaders have also urged the administration to cooperate with China to pace AI development.

Altman encouraged Trump to develop a pact with China that would address the risk of losing control over an AI agent, while ensuring that the safety measures do not leave either country at a disadvantage.

Ahead of Trump’s meeting with Chinese President Xi Jinping later this month, Altman appealed to the President’s very public desire for a Nobel Peace Prize.

“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 told Fortune magazine in an interview published Saturday.

“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,” he said.