AI Policy Wiki
Dashboard
⚲
AI Policy Wiki › People & Organizations

John Thune

medium confidence · updated 2026-09-26

US Senate Majority Leader (R-South Dakota); co-author with Amy Klobuchar of frontier AI legislation built on a duty-of-care liability standard rather than pre-market government review.

Contents

John Thune is the Republican senior senator from South Dakota and Senate Majority Leader.

Frontier AI legislation

Duty-of-care framework

Thune and Sen. Amy Klobuchar (D-Minnesota) are developing frontier AI legislation resting on a "duty of care" principle that would expose developers to legal liability for failing to manage catastrophic risks — regulating through litigation rather than pre-market review. The approach became public on August 3, 2026 in reporting that also identified the competing position held by Sen. Maria Cantwell (D-Washington), who has contemplated a pre-deployment regime under which the government would vet models before release. Commerce chair Ted Cruz (R-Texas) opposes giving the government power to block a release and favors the duty-of-care approach; the Thune, Klobuchar and Cruz offices did not respond to requests for comment on the account (Source: washingtonpost.com).

The Washington Post's account described the framework as resting on developer self-testing, and noted that AI companies are already subject to preexisting duties of care (Source: washingtonpost.com).

Klobuchar said at a July 30, 2026 Senate Commerce telecommunications subcommittee hearing that she and Thune aimed to produce a bill setting federal AI standards by the end of the August recess, telling the panel, "We can't just stand down and not have coordinated standards" on frontier models (Source: insideaipolicy.com). Neither the bill's text nor its position on preemption of state AI law had been made public as of early August 2026, and no measure sponsored by Thune on artificial intelligence appears in the Congress.gov legislation index for the 119th Congress as of September 1, 2026, the recess having ended (Source: congress.gov). The preemption question the bill would carry is covered at AI Federalism.

Negotiations and committee timetable

By September 2026 the negotiation had three principals rather than two: Thune, Klobuchar and Commerce chair Cruz, with Cantwell weighing in as the committee's ranking Democrat. Reuters reported on September 11, 2026 that the proposal would create a duty of care for AI developers and would apply to models with the most advanced capabilities, and that details remained under negotiation; Cruz wrote on X that he was working with Klobuchar and Thune to "address catastrophic risks involving biological or nuclear threats" (Source: reuters.com). Politico reported the same day that senators had not agreed on language and that the bill's path forward was unclear (Source: politico.com).

The unresolved question is who conducts the safety testing. A Democratic committee aide described the draft as having AI companies test their own models and present the results to the Commerce secretary for deployment approval, and characterized the Cruz–Klobuchar–Thune version as "primarily a voluntary standard type situation"; Cantwell's office, working to strengthen the testing regime, offered alternative language to Klobuchar's office, which the aide described as "very receptive," with Cruz's approval the deciding factor (Source: nextgov.com). A separate account of the draft describes it as carrying a clause that would override state AI law (Source: thenextweb.com).

A Senate Commerce markup of a version of the Thune–Klobuchar bill had been scheduled before the August 2026 recess and was cancelled, Cruz saying at the time that he wanted to focus on child-safety legislation (Source: politico.com). Cruz sought a markup on September 23, 2026, which would have required noticing it by September 16, and was pressing Klobuchar to introduce the draft; Klobuchar was reported to be wary of crossing Cantwell. The likely markup roster, according to one person familiar with the plans, included bipartisan legislation from Sens. Todd Young and Cantwell, the Future of Artificial Intelligence Innovation Act (S. 3952), which the Post described as formally establishing the Center for AI Standards and Innovation and creating public-private testbeds at the Department of Energy, the National Science Foundation and NIST (Source: washingtonpost.com; washingtonpost.com). In the bill text the testbed program is led jointly by NIST and the Energy Department in coordination with NSF, and foundation-model testing under it is voluntary (Future of Artificial Intelligence Innovation Act of 2026 — S. 3952 bill text (introduced February 26, 2026)). The legislative calendar constrained both chambers: House Speaker Mike Johnson did not intend to keep the House in session past September 17, 2026 despite calls from members of both parties to remain until AI legislation passed (Source: washingtonpost.com).

Thune's own public assessment in mid-September was guarded. On September 15, 2026 he told reporters he was continuing to work with Klobuchar on an AI bill that he thinks is the "right approach," adding, "But we'll see if, with everything going on right now … if it gets any traction"; aides said the unreleased bill builds on similar legislation the two advanced out of committee in the previous Congress. Majority Whip John Barrasso (R-Wyoming) said the same day that "I don't see a solution to the point that it's going to have a vote," pointing to the tight timeline and other priorities (Source: rollcall.com). On September 16, when Cruz said he would "love to mark it up this month, if we can get bipartisan agreement" and Klobuchar and Cantwell said a committee vote could come soon, Thune said it "remains to be seen" whether the proposal would get one; its introduction and possible committee action were still under discussion, and one person familiar with the talks named internal Senate politics, dynamics with the White House and support from the AI safety community as remaining obstacles (Source: semafor.com).

Safety advocates responded with minimum requirements for any Senate framework. A letter dated September 16, 2026, led by Americans for Responsible Innovation and signed by 20 organizations and five individuals, was addressed to Thune, Minority Leader Chuck Schumer and "Members of the U.S. Senate." Writing "As the U.S. Senate works to craft a framework for AI governance," it sets three minimum requirements: "we strongly request that any AI regulatory framework empower the government, not AI companies, to set mandatory, binding, and enforceable safety standards for advanced AI"; auditing of AI models that is "both mandatory and conducted independently"; and a federal floor that preserves states' authority to go further, as Congress "should not eliminate strong state laws in exchange for weaker federal standards." Signatories include Consumer Reports, Public Citizen, Public Knowledge, Common Sense Media, the Future of Life Institute, Encode AI, the Alliance for Secure AI, Issue One, the Institute for Family Studies, New York Assembly member Alex Bores and Daniel Kokotajlo (Senate coalition letter on redlines for AI safety legislation (Americans for Responsible Innovation, September 2026)). The letter names no bill; the Washington Post, which counted "19 advocacy groups" with AI researchers, reported it as a campaign to reject the legislation being developed by Thune and Klobuchar, and only the summary and quoted excerpts of that account were retrievable (Source: washingtonpost.com). It is distinct from an earlier letter of September 9, signed by 28 organizations, hosted by the same group and copied to Thune, which asked the President to publish the White House's frontier-AI review framework (Coalition letter to President Trump on public release of the White House frontier-AI review framework (September 2026)). Sen. Mark Warner (D-Virginia) said he was reviewing the bill but would not back it as a first step because "a lot of it's focused on after-action report, or incident disclosure after action" (Source: reuters.com). See Coordinated slowdown proposals compared.

In an interview at his Capitol office on September 23, 2026, Thune said he did not believe President Trump was "really dug in" against AI guardrails despite his public campaign against restricting the technology, having spoken privately with Trump as the issue became "pretty hot" a few weeks earlier; Thune reads Trump's position as a preference for companies, rather than Congress or the administration, coming up with ways to secure their systems, and some of Trump's pushback as directed at public "paranoia and noise around the subject." "I do think that the threats are real," Thune said, while adding that "there are people who are... doing a lot of fear mongering." He said Dario Amodei had called him to discuss AI safety and the bill before Amodei's latest public warning. Thune said the bill, which passed out of committee in the previous Congress, had not been reintroduced, and that Klobuchar was seeking more Democratic co-sponsors, including Chris Coons and Maria Cantwell, before reintroducing it. As he described it, the bill puts the onus on companies to mitigate AI risks and requires them to report serious risks from frontier models and plan to address them, and Democrats have two main objections: pre-deployment testing is something companies can opt into, where Democrats want it mandated, and Cantwell is asking for AI rulemaking authority at the Department of Commerce, which Thune worries could give the executive too much power (Source: axios.com).

Thune and Klobuchar had previously collaborated on the Artificial Intelligence Research, Innovation, and Accountability Act of 2023, which set out transparency and risk-management reporting requirements for companies deploying AI (Source: nextgov.com).

Separately, Thune said on September 14, 2026 that the Senate could take up data-center legislation sponsored by Sen. Jon Husted (R-Ohio) if the House passed it that week (Source: washingtonpost.com).

State attorneys general also addressed the congressional leadership. A letter dated September 23, 2026 from the attorneys general of 24 states, the District of Columbia and American Samoa to Thune, Johnson, Schumer and Jeffries asks for mandatory federal oversight of frontier-AI safety testing, government-led incident response and "explicit preservation of state authority by prohibiting the preemption of state laws" (Letter from state attorneys general to congressional leadership urging federal frontier AI regulation (September 2026)).

Open questions

  • The Klobuchar–Thune draft has been circulated to third parties but not published; its text, its preemption provision and its testing regime are known only through accounts by people familiar with the negotiations.
  • Whether a September 23, 2026 markup of other AI measures was noticed and held is not recorded here; Thune said on September 23 that the Thune–Klobuchar bill itself had not yet been reintroduced (Source: axios.com).

Relationships

Categories: People & Organizations | us | government | legislation | frontier-models | liability
What links here (38)