A sharp and increasingly fractious debate over how — and whether — to regulate artificial intelligence is reshaping American politics, pitting a deregulation-minded White House against a growing and ideologically eclectic coalition of voices urging caution. At the heart of the dispute lies a question that extends well beyond Washington's borders: can meaningful rules governing AI's risks coexist with the innovation that has become a significant engine of US economic growth?

President Donald Trump has staked out an unambiguous position, dismissing concerns about AI's dangers as a "hoax" and arguing that the technology represents "the greatest economic development engine in history." In a series of social media posts, Trump suggested that no formal guardrails are needed beyond his own authority, writing that "the only control or 'guardrails' that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT." His stance is reinforced by his informal AI adviser David Sacks, who has called the current wave of alarm a "panic" driven by electoral politics, and reflects the economic stakes involved: according to the forecasting firm ING, tech investment accounted for more than a third of all US economic expansion in the second quarter of 2026. Trump also frames AI leadership in explicitly geopolitical terms, describing it as a zero-sum contest with China — "whoever wins AI wins."

Yet dissent is coming from unexpected quarters. At the Pro-Human Assembly 2026, a conference organised by the nonprofit Future of Life Institute and held in Washington on 15 September, independent Senator Bernie Sanders of Vermont and former Trump adviser Steve Bannon — ideological opposites by virtually any measure — appeared back to back to call for urgent action. "I don't think you could find any two harder partisans," Bannon said. "And I think that speaks to the threat." Sanders, for his part, accused Congress under both parties of being "asleep at the wheel" and described Trump's position as "truly embarrassing." The event was catalysed in part by the resignation of Jacob Coxon, a researcher at AI company Anthropic, who publicly warned that the technology could pose an existential threat to humanity within years and accused companies of prioritising competition over safety. A July report from the United Nations had similarly cautioned that the gap between rapidly improving AI capabilities and effective risk management "may lead to catastrophic outcomes."

The proposals emerging from the assembly ranged widely — from mandatory transparency requirements and protections for whistleblowers to an immediate pause on the development of cutting-edge AI models, and even a permanent ban on superintelligence, a theoretical form of AI that could surpass human capabilities across a broad range of domains. Republican Representative Chip Roy of Texas called for light-touch guardrails, while Bannon advocated using presidential authority to force a slowdown. That breadth of opinion within the pro-regulation camp reflects a structural challenge: building consensus around AI governance is complicated by the technology's vast scope, touching everything from military systems to consumer applications.

For now, legislative action appears unlikely. Republican House Speaker Mike Johnson, who had earlier signalled openness to congressional oversight, retreated toward Trump's position this week, echoing the president's "hoax" framing. Analysts note that politicians on both sides are wary of alienating a public that polls show is sceptical of AI's promised benefits — but also reluctant to cross a president who has made AI a political touchstone. What may ultimately force the issue, observers warn, is not political will but circumstance. "It is possible that events in the future might electrify the public," said Matt Calkins, head of tech firm Appian. "And then positions — and talk — might not be so cheap."