House Republican leaders have determined that significant artificial intelligence safety legislation will not advance before the November 2026 midterm elections, leaving key proposals without a realistic path to a vote.

Speaker Mike Johnson and Majority Leader Steve Scalise have effectively shelved the most substantial bills addressing the risks of advanced AI, pointing to the need for further education among party members and a reluctance to take up complex legislation under pre-election pressure. As of September 2026, no major AI safety bill has reached the House floor — a holdup that leaves questions around AI risk disclosure and federal oversight without a near-term legislative answer.

Why Republicans Are Hitting the Brakes

Rep. Nathaniel Moran summed up the thinking, saying there are “three bad times” for AI policy: before an election, without education, and when driven by emotion.

Speaker Johnson has stressed that any AI policy must be meticulously crafted to protect US competitiveness, particularly against China. Earlier in 2026, House Republican leaders backed a White House AI framework centered on preempting state-level AI laws. But endorsing a framework is not the same as passing legislation. Because no such bill has moved, the preemption of state-level AI laws at the framework's core remains unrealized, leaving the boundary between federal and state authority over AI rules unsettled.

Adding to the headwinds, former President Trump has dismissed warnings about AI risks as a “hoax” — a stance that discourages Republican lawmakers who might otherwise champion safety-focused measures.

The Bills Gathering Dust

Two measures illustrate the stalled state of play. The bipartisan Great American AI Act, introduced in July 2026 by Reps. Jay Obernolte and Lori Trahan, would require AI developers to disclose risks and would give the Commerce Department authority over catastrophic risks. Without leadership backing, the bill sits in legislative limbo.

Rep. Moran's AI Incident Reporting Act, a narrower proposal aimed at transparency around AI safety incidents, has likewise stalled. Until either reaches the floor, the transparency each pursues — developer risk disclosures and formal incident reporting — remains a proposal rather than enacted policy.

Congress is now entering a seven-week recess, during which discussions are expected to continue informally. The earliest realistic window for floor consideration is November, after the midterm elections, and even that session offers only a compressed runway. Under standard congressional procedure, any bill not enacted by the time the new Congress is seated in January 2027 expires and must be reintroduced from scratch, meaning these measures would need to clear the House within weeks of lawmakers' return or begin the legislative process anew next year.

This article is based on reporting from CryptoBriefing.