The Senate’s AI Question: Who Builds the Guardrails, and Where?
The Senate’s AI Question: Who Builds the Guardrails, and Where?
Senate Democrats are debating how to set up AI rules when they take the majority in January, and the fight is as much about process as policy. As reported by the Associated Press, Colorado Sen. John Hickenlooper has proposed a new select committee on artificial intelligence, a move that would pull the issue away from the committees already working on it.
The Endorsement Holdout
Hickenlooper has withheld his endorsement in the race for Senate majority leader between Chuck Schumer and Maryland Sen. Chris Van Hollen. Instead, he wants whoever wins to present a “day one plan for responsible regulation” of AI. His reasoning, as noted in the AP’s reporting by Mary Clare Jalonick, is that Democrats formed a working group on the issue three years ago and have made too little progress since. Republicans have led both chambers for the past two years and have talked up AI as a priority, but their focus has been mostly on innovation, with no major push for safeguards or monitoring.
“There’s a perception that we haven’t done anything,” Hickenlooper said.
“There’s a perception that we haven’t done anything.” — Sen. John Hickenlooper
What Democrats Want to Tackle
Democrats have been discussing possible first steps, which could include regulating data centers, setting safety standards, and expanding monitoring and research. Hickenlooper’s own priorities are robust third-party evaluations, measures to address national security risks, and “clear rules of the road to hold AI companies accountable.” Schumer’s office says he has been working with the incoming committee chairs to build an “ambitious AI agenda,” and Schumer is favored to win the closed-door leadership vote on Dec. 1.
A Committee Built for an Election Clock
Hickenlooper’s pitch is organized around timing. In his letters to both Schumer and Van Hollen, he described the next two years as an audition: “With the presidential election two years away, we have a two-year audition to show Americans we can deliver on three priorities: affordability, accountability, and AI safety.”
Van Hollen backs the idea of a joint committee, arguing that “AI covers so many different areas that are impacting our lives.” That breadth is also the problem. Committee chairs who are already drafting AI proposals have not signed on. Sen. Ron Wyden, the incoming Finance Committee chair, wants to “see the specifics” before deciding. Sen. Bernie Sanders, expected to chair the HELP Committee, has not endorsed the idea yet, though he insists Democrats must prioritize the issue “no matter what particular way we end up approaching it.”
“Congress has been way, way, way behind where it should be in protecting the American people.” — Sen. Bernie Sanders
Republicans Push Back
Texas Sen. Ted Cruz, the current Commerce Committee chairman, says moving the issue out of his committee and others “makes no sense.” He still sees room for cooperation, pointing to bipartisan conversations already underway and his hope that the next Congress can find “common ground on addressing catastrophic risk.”
A Record of Little Action
The federal track record is thin. President Joe Biden’s administration issued a 2023 executive order on AI oversight, which President Donald Trump rescinded soon after returning to office. A bipartisan working group formed by Schumer released a 2024 report recommending at least $32 billion over three years for AI development and safeguards. Trump, who holds the veto, has repeatedly argued that little regulation is needed.
What Comes Next
House Democrats have their own robust package of proposals, and three of them recently launched a House AI Safety and Innovation caucus focused on security risks. Whether the Senate builds a new committee or spreads the work across existing ones may decide how quickly any AI rules can move.
Aaron Rose is the founder and publisher of Tech Reader Magazine, covering system architecture, cloud platforms, and AI policy.