Ted Cruz blocks the Warner–Schatz–Kim AI Risk Management and Security Act, which would give a federal AI Safety Board 45 days with frontier models before release
On Sept 29, 2026, the day of the White House frontier-AI accord, Sen. Ted Cruz objected when Democrats Mark Warner, Brian Schatz and Andy Kim asked for unanimous consent to pass their Artificial Intelligence Risk Management and Security Act of 2026 (introduced Sept 24). The bill would create a permanent AI Safety Board in the Commerce Department and require developers of covered frontier models to give it access to weights and runtimes at least 45 days before public release, with civil penalties of up to $250,000 per violation per day.
Key facts
- Introduced Sept 24, 2026 by Sens. Mark Warner (D-VA), Brian Schatz (D-HI) and Andy Kim (D-NJ)
- Permanent AI Safety Board in the Commerce Department with NIST, Commerce, CISA, NSA and Treasury representatives plus independent technical experts
- Developers must give the board access to qualifying models at least 45 days before release, including weights, configuration files, runtimes and software libraries
- Civil penalties up to $250,000 per violation, each day counted separately
- Incident reporting within 30 days (72 hours for imminent threats), a national AI incident database run by NIST and CISA, and standards for autonomous agents' data access, tool use and authority levels
- Sept 29: Cruz objected to unanimous consent, saying he has 'real reservations about this specific proposal' because it lets the executive branch pick the board that sets the rules, while offering to work on other approaches (The Hill)
- Warner: 'The companies building the most powerful AI systems are warning us that the technology is advancing faster than our safeguards'
What happened
Warner, Schatz and Kim went to the Senate floor to try to pass their new bill without a vote. Cruz, who chairs the Commerce Committee and is negotiating a separate leadership bill, objected. His concern was the discretion it gives the executive branch, not pre-release testing as such.
Why it matters
It was the second unanimous-consent attempt on an AI-safety bill blocked in two weeks (after Kennedy's kill-switch bill on Sept 17), on the same day the White House chose a voluntary accord with six labs. The bill's 45-day pre-release access with weights goes further than any enacted US rule.
Changelog
- 2026-10-03: created
Related events
- Thune, Cruz and Klobuchar negotiate a Senate bill imposing a 'duty of care' on frontier AI developers and letting the government block unsafe model releases ★★★★
- Sen. John Kennedy's AI 'kill switch' bill is blocked in the Senate after Rand Paul objects ★★
- Trump hosts AI CEOs at the White House; they sign a voluntary 'morally binding' Accord on Superintelligence, and Trump rejects new federal AI rules ★★★★
- Reps. Trahan and Obernolte introduce the bipartisan FRONTIER Act: licensed independent auditors, incident reporting and state preemption for frontier AI ★★★
- Senate subcommittee holds first hearing on rogue AI agents; Hawley pushes developer liability after Altman declines to testify ★★★★
Sources (6)
- pressThe Hill: Cruz blocks push by Democrats to unanimously pass AI safety bill
- officialSen. Warner: On Senate floor, Warner discusses urgent need to pass AI security legislation
- officialSen. Schatz: Schatz, Warner to take to Senate floor to demand passage of new AI security legislation
- pressFalls Church News-Press: Warner warns AI risks 'no longer theoretical' (Sept 29)
- pressDefend Intelligence: Warner, Schatz and Kim introduce a bill on advanced AI models
- pressImplicator: Cruz blocks AI bill requiring 45-day model access as Trump backs voluntary accord
id: 2026-09-29-cruz-blocks-ai-risk-management-security-act · updated 2026-10-03 · open in the interactive timeline