AI safety warnings aren’t only coming from the left, despite Trump’s dismissal

President Donald Trump used his Sept. 22 speech to the United Nations General Assembly to dismiss concerns about artificial intelligence, equating them to other issues he considers “hoaxes.”

“The very same people who said we’ll all be dead in 12 years because of global warming, a name since reborn to climate change because the planet was cooling, not warming, and nobody was dead,” he said. “These are the same people that are now saying that AI is going to kill us all, that robots are going to attack us, and that everything is going to be a total disaster. Same group of people. This is the group that came up with the Russia, Russia, Russia hoax, the Ukraine, Ukraine, Ukraine hoax, climate change, open borders.”

Trump also equated AI fears to a “hoax” in recent Truth Social posts. 

Warnings that AI could wipe out humanity or do significant harm gained renewed attention after Jacob Coxon, a researcher from leading AI company Anthropic, resigned in September, publicly citing such an existential threat.

A White House spokesperson told PolitiFact in an email that Trump was referring to “the numerous Democrats who are currently echoing AI hysteria and were or still actively are pushing hysteria related to climate change and the Russia collusion hoax.” 

Although some AI regulation advocates may overlap with those concerned about climate change, worries about AI are not confined to one group of people or one political party. The warnings even come from the top of the industry itself.

On Sept. 12, Anthropic CEO Dario Amodei published an essay calling to slow down AI development. OpenAI CEO Sam Altman, Google DeepMind Cofounder and Chair Demis Hassabis and xAI founder Elon Musk echoed his concerns — and agreement among those leaders is rare.

Robin Jacobson, a University of Puget Sound politics professor, said AI safety is an issue not currently owned by either political party, and the movement is comprised of “strange bedfellows.” 

“The public, across parties, has real growing concerns with AI,” Jacobson said. “So I think, in fact, you’re going to see, as opposed to climate change — where this was about convincing the public there was this looming threat — I think the public is actually demanding action on concerns.” 

Scientists widely agree that climate change is real, is driven by human activities and is causing the Earth’s temperature to warm. The United Nations Intergovernmental Panel on Climate Change in 2018 pinpointed 2030 as a benchmark for dealing with climate change, but it didn’t say humans would all die in 12 years.

What Trump referred to as the “Russia hoax” is the investigation led by Special Counsel Robert Mueller into possible connections between Trump’s 2016 campaign and Russia. Democrats publicized and attacked Trump over the investigation, but it continued under Trump and his deputy attorney general Rod Rosenstein appointed Mueller special counsel, so the scrutiny didn’t come only from Democrats. 

Trump described his first impeachment in 2019, for attempts to pressure Ukrainian President Volodymyr Zelenskyy to announce an investigation of former Vice President Joe Biden ahead of the 2020 election, as a “hoax.”

Other than elected Democrats who supported the investigations and are now calling for AI safety, there isn’t a coordinated group that was behind both the investigations and AI safety warnings. 

Recent events caused growing alarm from both sides of the political aisle

Recent warnings from researchers about losing control of AI come as companies have had their AI systems take action without human authorization. Researchers say incidents such as OpenAI’s agents attacking the company Hugging Face show that autonomous AI agents can take harmful actions that aren’t the intended task, a problem called “misalignment.” 

Researchers have warned that AI systems are advancing faster than their creators’ ability to reliably understand their behavior and contain them.

In experiments, researchers have documented AI agents scheming to conceal information from human monitors, cheating on tests to receive a higher score and attempting to disable monitoring. AI agents are sometimes aware they are being evaluated. As AI handles more of the research that goes into the next generation of AI, experts say those behaviors could become harder to detect. 

Lawmakers from both parties have announced or introduced legislation to regulate AI, including some that address advanced AI that could cause widespread harm. Sen. Bernie Sanders, I-Vt., and Rep. Greg Casar, D-Texas, announced legislation to ban artificial superintelligence. “Superintelligence” is generally understood by industry experts as an AI system that is far ahead of human intelligence at nearly all cognitive tasks, including scientific research and communications.

Rep. Ted Lieu, D-Calif., and Rep. Nathaniel Moran, R-Texas, introduced the “AI Kill Switch Act,” which would require AI companies to have mechanisms to shut down AI systems that can “cause catastrophic harm.”

Some conservative lawmakers have been outspoken in warning against AI. 

Sen. Josh Hawley, R-Mo., launched an investigation in September into AI’s existential risk. In a letter to Altman, he cited concerns of AI killing all humans in the near future. 

Rep. Anna Paulina Luna, R-Fla., said on Fox News that there should be a kill switch to “protect humanity.” 

Former Trump adviser Steve Bannon has called for regulating AI, even joining Sanders in a “pro-human” gathering on AI.

On the political left, the AI fears are not universal. The New York City Democratic Socialists of America chapter dismissed warnings of AI catastrophe as “AI alarmism and AI hype,” Wired reported. 

Among Americans, those who believe there’s at least a “moderate risk” of AI destroying humanity covered majorities of both political parties, according to a September Politico poll. Among those polled who voted for Democratic candidate former Vice President Kamala Harris in the 2024 presidential election, 70% believe there is a risk of AI destroying humanity, compared with 61% of those who voted for Trump.

Daniel Schiff, who co-directs Purdue University’s Governance and Responsible AI lab, said AI policy in general doesn’t fall cleanly along partisan lines. Issues such as data centers, privacy concerns, global competition and job displacement activate different groups of voters across parties. 

“It’s fair to say the parties have not really crystallized, even if there are some prominent figures who are taking stronger stances recently,” Schiff said. 

There is some overlap between climate change and AI safety advocates, including among a loose movement called effective altruism, which is focused on finding the best ways to help the most people. 

People in that movement have long considered both climate change and artificial intelligence risks to be serious problems. But it’s not the only group concerned about loss of control of AI.

Our ruling

Trump said the people who warned of climate change, the “Russia hoax” and the “Ukraine hoax” are the same people “now saying that AI is going to kill us all.”

While advocates for AI regulation and safety have some overlap with climate activists and Democrats who participated in those investigations, they are not confined to a singular group. Politicians from both parties have spoken out or announced legislation based on concerns about the capabilities of advanced AI. 

We rate this claim False.

RELATED: Fact-check: President Donald Trump misleads on Iran war, AI in speech to UN