AI Safety Divides the American Left as Democrats Debate Present and Future Risks

September 10, Two days after Jacob Coxon, a researcher who recently left Anthropic, warned that the artificial intelligence industry could spell the end of humanity, the Democratic Socialists of America’s New York City chapter published an image carousel on Instagram.

The first slide showed an ouroboros—a dragon eating its own tail—against a hot pink background. “AI alarmism and AI hype are the same story. Both draw attention to a sci-fi future to distract from the real harms we are currently experiencing,” the image read.

The post was written by the New York City DSA’s Tech Action Working Group, which focuses on technology regulation for the chapter, the largest in the country. It received more than 25,000 likes on Instagram, but not everyone agreed with its message. One commenter called it “a horrifying and dangerous view,” while another wrote, “Sometimes sounding the alarm is justified.”

Why AI safety is dividing the American left

An ideological divide is emerging on the American left as politicians try to understand the risks of AI. At the center of the debate is whether to accept technology companies’ claims that their products could become powerful enough to threaten humanity.

Some on the left question the value of elevating theoretical concerns about AI over harms that are already affecting people. They argue that repeating warnings promoted by frontier AI labs could empower the companies developing the technology, distract from current problems and weaken efforts to regulate those harms.

“AI has become the latest tool of capitalist exploitation, and the damage that unregulated and misused AI is causing is already evident across the country,” Alan Dunn, a member of Tech Action, told WIRED. “As socialists, we don’t trust these companies to create their own regulations.”

Others are following the lead of Sen. Bernie Sanders, the unofficial leader of the American left, who has forcefully raised concerns about AI safety and argued that AI could potentially wipe out humanity. They say policymakers can address current harms and future existential risks at the same time.

“The same people who are concerned about the development of frontier governance and superintelligence are also grappling with the current evils,” said Alexander McCoy, a Democratic strategist who works on AI issues. “We need to address all of these important factors. These cutting-edge AI capabilities are real and their power is increasing. They are not fake or hype.”

Bernie Sanders brings AI risks to Washington

Over the past year, Sanders has immersed himself in the AI safety debate, bringing a conversation previously concentrated in Bay Area technology circles to Washington and his large populist base.

The 85-year-old senator was an early participant in two major left-wing organizing efforts this year focused on data centers and the potential harms of AI, including job losses and existential threats. His data center moratorium bill, announced in March, explicitly linked construction pauses to the need for federal regulations that would protect “the future of humanity.”

Sanders’ position has also placed him in an unusual alliance, most famously with Steve Bannon, a former adviser to President Donald Trump and a right-wing populist podcast host. The connection has also influenced a wave of messages from left-wing candidates running across the country, many of whom have linked broader concerns about AI to opposition to data centers.

Left-wing candidates call for AI regulation

On September 9, the day before the New York City DSA post, six candidates from the Working Families Party—including four supported by Sanders—released statements calling for Washington to regulate AI and prevent threats to humanity’s survival:

• Candidate statement
• Candidate statement
• Candidate statement
• Candidate statement
• Candidate statement
• Candidate statement

The growing debate highlights a broader question for the left: whether AI’s immediate social and economic harms should take priority over warnings about a potentially catastrophic future—or whether both concerns demand action now.