Congress is once again delaying action on artificial intelligence, pushing the debate past the November midterms and leaving lawmakers who want guardrails increasingly frustrated. With the House out of session and the Senate facing a packed schedule before a monthlong recess, the prospect of comprehensive AI legislation this year is fading—even as public concern about the technology reaches new heights.
Members from both parties argue that Congress must move faster on AI safety, national security, and the technology's impact on critical infrastructure. But Republican leaders have cautioned against heavy-handed regulation that could stifle innovation or weaken the U.S. in its rivalry with China. Speaker Mike Johnson (R-La.) and President Trump are scheduled to meet with AI industry leaders next week, yet legislation is likely to be postponed until a post-midterm lame-duck session—or even the next Congress.
“I know we have been trying, both Democrats and Republicans, to get Mike Johnson to allow us to actually bring it up in this Congress. I don’t know whether it will happen in the lame-duck session,” Rep. Don Beyer (D-Va.), co-chair of the bipartisan Congressional Artificial Intelligence Caucus, told The Hill. “My guess is, if the president still thinks it’s a hoax, that Mike Johnson won’t bring it to the floor, won’t make it a priority. If Donald Trump suddenly gets the insight that federal regulation of AI is a good thing, then yes, we could. We have a couple of weeks there, and we can actually do something meaningful.”
Johnson has previously advocated for a “balanced approach” to regulation, warning that “the reflex of legislative bodies is to cover things up with red tape and in hyperregulation.” He added: “If you do that, then you lose your edge, and that is a national security concern for the American people.” Trump, who wants AI rebranded as “super intelligence,” has consistently opposed new guardrails.
Next week’s meeting, which Johnson pushed for, comes amid a flurry of bills introduced on Capitol Hill. Sen. Bernie Sanders (I-Vt.) and Rep. Greg Casar (D-Texas), chair of the Congressional Progressive Caucus, unveiled the Ban Artificial Superintelligence Act, which would prohibit the development or deployment of artificial superintelligence and create a cabinet-level Department of Artificial Intelligence to protect the public. “When you are racing towards a cliff, you don’t just ease up on the gas pedal. You hit the brakes,” Sanders said. “When the future of humanity is at stake, we cannot let a handful of Big Tech CEOs write their own rules. Scientists have been clear: AI is an existential threat to humanity. It must be treated as such.”
Other proposals include the bipartisan Stop Rogue AI Act from Reps. Mike Lawler (R-N.Y.) and Josh Gottheimer (D-N.J.), which aims to help businesses and federal agencies detect and stop dangerous AI agents in their networks. In July, Reps. Nathaniel Moran (R-Texas) and Ted Lieu (D-Calif.) introduced the AI Kill Switch Act, requiring developers to maintain the ability to throttle, suspend, or shut down their systems. This month, Sen. Rand Paul (R-Ky.) blocked a request by Sen. John Kennedy (R-La.) to pass legislation mandating a “kill switch” for superintelligent AI. Kennedy’s bill would have put companies, not the government, in charge of using an emergency switch. Paul warned that acting hastily “before the technology is understood” risks “killing innovation and setting this technology back decades.”
Senate Majority Leader John Thune (R-S.D.) told Axios that AI threats are “real” and that “you have to have at least some sort of a plan and some guardrails around it.” He pointed to an AI safety bill he introduced with Sen. Amy Klobuchar (D-Minn.) that lawmakers are negotiating. Still, time is short: Republicans will have less than two months after the midterms to strike a deal before the new Congress is sworn in—a window that could narrow further if Democrats win the House.
Beyer said he expects Democrats to work in “good faith” with Republicans and Trump on any AI regulation. “It doesn’t do us any good to pass a House bill that dies the day it hits the Senate. So there are a lot of people in the Senate again from both parties that are eager to do something that shows that we’re acting responsibly on AI,” he said. “Even if we have differences in opinion, we need to try to narrow them as much as we can. We don’t want to just pass a bill for the sake of saying, ‘Look, we did something good in the House.’ We want to make sure it becomes law.”
As the debate drags on, related issues are piling up. The Energy Department recently announced a $2 billion plan to modernize the aging power grid to meet AI’s soaring energy needs. Meanwhile, some critics argue that industry-led safety efforts, such as those from OpenAI’s Sam Altman, amount to a push for a regulated monopoly. And on the healthcare front, the Trump administration’s move to strip 760,000 from ACA rolls has sparked its own affordability battle, underscoring the broader policy logjam.
