Sanders and Bannon Agree: AI Must Be Stopped
When your enemies agree, pay attention
The political divide in America has rarely looked wider. Every issue -- from economics to immigration to foreign policy -- splits the country into two camps that do not speak the same language.
This week, the two men who embody those camps more vividly than anyone else stood on the same stage and delivered the same message.
Bernie Sanders and Steve Bannon both told a Washington audience that artificial intelligence must be slowed before it is too late.
"We are racing towards a cliff," Sanders said. "When you are racing towards a cliff, you hit the brakes."
Bannon went further. Legislation will not work, he argued. Only a presidential executive order can stop what he called "the oligarchs" from building a future Americans did not consent to. "This is a hinge in history," he said.
If the two ends of the American political spectrum found exactly one thing to agree on, it is worth asking what made that possible -- and what it means for an industry that has spent two years assuming regulation would never arrive.
The Pro-Human Assembly
The occasion was the Pro-Human Assembly, hosted in Washington by the nonprofit Future of Life Institute. Several hundred attendees listened to an extraordinary lineup that included members of Congress, the Archbishop of San Francisco, the actor Ashley Judd speaking for SAG-AFTRA, and parents whose teenagers died by suicide after forming relationships with AI chatbots.
Hats distributed at the event carried a slogan that summed up the gathering's thesis: "Keep the future human."
Representative Greg Casar, Democrat of Texas, put it bluntly: "We ban AI systems that are too powerful for humans to control." Representative Lori Trahan, Democrat of Massachusetts, said: "I can't believe I have to say this. The future we all want is human. It is nonnegotiable."
The religious leaders present framed the issue in moral terms. Salvatore Cordileone, the archbishop of San Francisco, warned: "We have to be the masters of AI."
A coalition from opposite extremes
The remarkable thing about the Pro-Human Assembly was not that anyone wanted AI regulation. The remarkable thing was who wanted it together.
Sanders, the democratic socialist from Vermont, has spent his career fighting corporate power. His proposed solution is a data center construction freeze and a deal between President Trump and China's Xi Jinping to jointly ban the development of superintelligent AI. It is the kind of international arms-control approach usually reserved for nuclear weapons.
Bannon, the architect of Trump's 2016 victory and a figurehead of the nationalist right, arrived at the same destination from the opposite direction. His concern is not corporate power but democratic consent. "American citizens are not going to be supplicants to the oligarchs anymore," he said.
The two men disagree on almost everything. That they agree on this is the most politically significant signal the AI industry has received all year.
Meanwhile, OpenAI backs the other path
While the activists debated executive orders and construction freezes, a quieter development in the same city showed that the industry itself is preparing for regulation.
OpenAI told sponsors of the FRONTIER Act -- a bipartisan bill introduced by Republican Jay Obernolte and Democrat Lori Trahan -- that it could support the bill's key provision: mandatory independent safety audits for frontier AI labs.
Chris Lehane, OpenAI's head of global affairs, confirmed the company's position. It is the first time OpenAI has backed a concrete federal mandate for outside safety auditors. The company had previously supported a California state bill and CEO Sam Altman had agreed to voluntary outside audits in response to a challenge from Anthropic CEO Dario Amodei. But endorsing federal legislation with teeth is a different step entirely.
The FRONTIER Act would create the first federal framework for AI safety, requiring labs above specific revenue and compute thresholds to submit to independent verification organizations (IVOs). The bill does not dictate what the auditors must find -- but it forces the labs to let them look.
The White House stands alone
The pro-regulation coalition growing in Washington is notable in part because it leaves one major figure isolated: President Trump, who has called fears about AI a "hoax" and threatened AI executives with criminal prosecution for harms caused by their systems.
Trump's position puts him at odds not only with Democrats and religious conservatives, but also with his former strategist Bannon and with an industry that is increasingly asking to be regulated rather than left to manage existential risk on its own.
House Democrats have called for canceling the upcoming congressional recess to legislate on AI. Senator Elizabeth Warren has backed a pause. Senator Richard Blumenthal warned the country is "on the verge of losing control." Ted Lieu, the No. 4 Democrat, is pushing for a committee vote on a bipartisan AI bill. Even the so-called "Godfather of AI" has said tech regulation is nearing a "Covid-style pivot moment."
The political consensus on AI is not there yet. But it is forming faster than anyone predicted -- and it is coming from both ends of the spectrum at once.