U.S. enacts AI safety bill before 2027?

Prediction markets put the probability at 12%: U.S. enacts AI safety bill before 2027. Currently, markets see this as unlikely (12% YES).

What’s Happening

The probability that the U.S. enacts a federal AI safety bill before 2027 remains low at 12%, even as state-level momentum accelerates. On July 6, 2026, Illinois Governor J.B. Pritzker signed the Artificial Intelligence Safety Measures Act (SB 315), making it the third state after California and New York to impose binding frontier-model regulations. The Illinois law requires covered developers to implement catastrophic-risk management, safety frameworks, and incident-reporting protocols, with stricter duties for "large frontier developers." Connecticut followed a month earlier, with Governor Ned Lamont signing Public Act No. 26-15 on May 27, 2026, which adds whistleblower protections for frontier model employees and mandates provenance labeling for generative AI content. These actions create a patchwork of compliance obligations that industry groups have long warned could trigger federal preemption efforts [Norton Rose Fulbright, Jul 22].

The push for a single national framework faces significant headwinds from the executive branch. President Trump’s December 2025 executive order directed the Department of Justice to challenge state AI laws deemed inconsistent with federal policy, specifically naming Colorado’s SB 205 as a target. That order, combined with the administration’s preference for voluntary guidelines over binding statutes, has kept the U.S. enacts AI safety bill scenario largely stalled in Congress. Meanwhile, the New York Responsible AI Safety and Education (RAISE) Act, enacted on December 19, 2025, imposes transparency duties on developers whose training compute costs exceed $5 million per model and $100 million in aggregate — thresholds that capture only the largest labs. Legal analysts note that Connecticut’s employment and frontier model provisions could face scrutiny under the DOJ directive, even as its child safety and procurement sections fall within recognized carve-outs [Freshfields, May 05].

What happens next hinges on whether Congress can reconcile competing priorities before the 2027 deadline. The current 12% probability reflects the reality that no comprehensive federal bill has cleared committee in either chamber, despite repeated hearings on AI risk. State laws now cover roughly 15% of the U.S. population, creating pressure on multinational firms to comply with divergent rules. Industry observers point to the possibility of a narrow federal bill addressing only catastrophic-risk reporting or election deepfakes, which would be a far cry from the comprehensive safety regime envisioned by earlier proposals. The DOJ’s pending litigation against Colorado, expected to reach trial in early 2027, could set a precedent that either validates state authority or clears the path for federal preemption — a ruling that would directly influence whether the U.S. enacts AI safety bill legislation in the remaining window [JD Supra, Aug 07].

Traded on Polymarket — $102K Volume

Polymarket prices this at 10c YES with $102K in volume. Moderate liquidity — use limit orders for positions above $1K to avoid moving the price.