NEW YORK (AP) — Amid growing anxiety over the risks posed by rapidly developing artificial intelligence technology, potential Democratic candidates for president are rushing to show that they're taking the warnings seriously — while Donald Trump brushes off concerns — and acknowledging they need a reluctant Washington to step up.
On Friday, California Gov. Gavin Newsom signed an executive order to accelerate implementation of a California law that
calls for independent oversight of AI companies and the use of “kill switches” so developers have the ability to shut down their systems.
“California will do what it can. But this nation needs to get in the game. Washington, D.C. needs to do its job,” Newsom said on his podcast Friday.
The order came a day after Pennsylvania Gov. Josh Shapiro, also expected to be a leading Democratic presidential prospect, delivered a speech outlining the steps he’s taking locally to confront AI risk. Like Newson, he conceded that states have obvious limitations in regulating a global technology.
The efforts come as AI leaders have issued dire warnings about the potentially catastrophic threat posed by rogue AI agents. The country has begun grappling with the impact of a technology that could bring both tremendous gains in areas like medical research and automation but also unleash deadly viruses, cause widespread economic upheaval and help terrorists inflict harm.
While Trump has angrily dismissed the warnings as a “HOAX,” Democratic lawmakers and candidates across the country are delivering speeches and proposing legislation as they sense a political opening heading into the midterms and the next presidential race.
Anthropic’s Dario Amodei, OpenAI’s Sam Altman and SpaceXAI’s Elon Musk have all called for greater government oversight of the industry after a series of hacks revealed AI agents conspiring to deceive humans.
Trump has responded by saying that he is “the only control or ‘guardrails’ that AI needs.”
Democrats quickly pounced.
Georgia Sen. Jon Ossoff, a Democrat who is seeking reelection in a state Trump won two years ago, said at a campaign rally last weekend that “a capable president would move swiftly to secure and reassure the American people, rush inspectors into frontier labs, fortify the nation against bioterrorism, demand legislation from Congress, and lead the world toward an AI treaty.”
Ossoff, seen as a potential 2028 contender, said geriatric politicians have little comprehension of AI technology.
“An ancient establishment, barely capable of email, sleepwalks into a technological revolution while the president dismantles precautions and calls us fools for worrying,” he said.
Another potential White House contender, Arizona Sen. Mark Kelly, told Politico he will introduce legislation next week to create a fund that AI companies would contribute to support workers affected by the disruption.
Shapiro, in his speech, called for new federal rules to require independent, third-party oversight, increased transparency, and to “prioritize safety above all else.” In his state, he noted that he has already signed an executive order to establish “real guardrails governing how Pennsylvania uses generative AI.” But he acknowledged that states have obvious limits.
“It’s going to require national and, in fact, international, leadership and collaboration,” he said.
Alex Bores, a soon-to-be-former New York State lawmaker who lost his congressional primary this summer after AI companies angry over an AI safety bill he championed spent millions to defeat him, said: “The salience of this issue has climbed dramatically.”
“People want there to be real action on it,” he said. “Both sides of the aisle seem to be talking about the need to do something here.”
Bores, who recently launched a political advocacy group aimed at influencing how his party responds to the technology, said he was encouraged by governors who seemed to be pivoting from calling on the federal government to act to working on their own regulations.
“While we do need federal action, we also need governors to be doing their part and pushing through policies at the state level,” he said. “I’ve been encouraged by the vibe shift” over the last few days.
While it remains early, polling suggests a growing ideological divide over AI, with Democrats becoming more alarmed over the last few years, while Republicans have become less concerned.
Democrats are more likely than Republicans to be “more concerned than excited” about the growing use of AI in daily life, according to a June Pew Research Center poll.
About half of U.S. adults overall say the increased use of AI in daily life makes them “more concerned than excited.” Another Pew Research Center poll from June found that about 7 in 10 U.S. adults think that AI will cause job losses in the U.S. over the next 20 years, up 7 percentage points since 2024.
Arizona Sen. Ruben Gallego, a Democrat, has been calling on Republican Senate leadership to establish a select committee on A.I. as soon as possible.
“I worry about this as a true existential threat. This is the first time that we’ve created technology to destroy humanity that could be controlled by not humans,” he said on MSNOW.
While potential 2028 candidates have been eager to weigh in on AI, the issue has been less potent in many local races, where some fear the kind of spending that Bores faced.
At a rally in New York's highly competitive 17th Congressional district on Wednesday, the audience cheered as Democratic candidate Cait Conley vowed to ban data centers that power AI and cloud computing. But no direct mention was made of the broader technology or its risks.
There have been exceptions.
In Michigan, both candidates for a tightly contested Senate seat have called for some level of slowdown. At a town hall on Sept. 11, Democratic nominee Abdul El-Sayed said AI was keeping him up at night, describing it as an existential risk.
“We need to all hit pause,” said El-Sayed, who wants to halt frontier research and mandate new public oversight, including turning AI companies into public benefit corporations, creating a new regulatory agency and banning the use of AI to deny medical care, make hiring or firing decisions — or fire a weapon.
On Monday, Republican nominee Mike Rogers called for an AI slowdown in an X post after Amodei’s warning.
“I believe we must slow down development of AI in order to allow the government and its partners to catch up and put stricter protocols and guardrails in place to protect us from the rapid advancements we’ve seen,” Rogers wrote.
Bores said that data centers had fueled such backlash not only because of concerns over their immediate local impact, but also because, “It is the first time that AI is meeting democracy.”
“Of course they're channeling all of their concerns about it into the fight against data centers,” he said. ____
Steve Peoples in New York, Linley Sanders in Washington and Jacqueline GaNun in Lansing, Michigan contributed to this report.
