What happened
On 29 September 2026, Sen. Ted Cruz (R-TX) objected on the Senate floor to a unanimous-consent attempt by Sens. Mark Warner (D-VA), Brian Schatz (D-HI) and Andy Kim (D-NJ) to pass the Artificial Intelligence Risk Management and Security Act of 2026. The bill would create a permanent AI safety board within the Commerce Department with power to develop enforceable frontier-model evaluation standards, require developers to give the panel access to models at least 45 days pre-release, and levy fines for noncompliance. Cruz cited concerns over granting the executive branch broad, arguably undefined authority over private AI companies.
Why it matters
This was the highest-profile federal legislative attempt in this window to impose binding pre-deployment government safety review on frontier AI models, and its floor failure removes near-term prospect of a mandatory US frontier-safety review regime. It underscores that US federal AI safety legislation remains stalled, leaving state AGs, FTC, and states like California as the operative enforcement venues — a key signal for AI deployers building compliance programs around expected federal mandates.
Action needed
No direct compliance action, but organisations tracking US federal AI regulation should not assume binding frontier model pre-deployment review will pass imminently; weigh state and FTC enforcement risk accordingly while Senators say they will continue to press the bill through regular order.