U.S. enacts AI safety bill before 2027?
The odds still look low that Congress will enact a federal AI safety bill with the required restrictions before the end of 2026. There is real bipartisan interest and some committee-level movement, but the calendar, unresolved policy disputes, and lack of floor momentum all point toward a likely miss.
Analysis
The current setup strongly favors No because the remaining path to enactment is narrow and time is short. The House has effectively paused major action until after the election cycle, several AI safety proposals remain trapped in committee, and the Senate still has not settled on final bill language. Even where there is broad agreement that AI deserves federal attention, that is not the same as agreement on a bill that survives bicameral passage and reaches the President before the deadline. The market price around 12.5% for Yes looks directionally sensible, and if anything the news flow slightly reinforces the view that a near-term federal law is more likely to slip into 2027 or later than to be finished this year.
The legislative process itself is the main obstacle. A bill that qualifies for this market must include at least one substantive restriction such as limits on model creation or release, training constraints, usage restrictions, or human-in-the-loop requirements, which are exactly the kinds of provisions most likely to trigger disagreement. House committee progress does show that lawmakers are drafting and refining ideas, but committee passage is still far from floor passage, and floor time is scarce. In the Senate, bipartisan negotiations have been discussed for months without final agreement, and that signals that the core tradeoffs are still unresolved. With the House away, the election approaching, and the post-election window compressed by holidays and end-of-session priorities, the procedural probability of completion remains low.
Historical and political context also cuts against Yes. Congress has repeatedly struggled to convert AI concern into durable federal regulation, while states have moved faster and more aggressively, which reduces the urgency for a quick federal compromise. The split between safety advocates and those pushing preemption or lighter-touch rules makes a clean deal harder, because any bill must satisfy enough lawmakers in both chambers and likely withstand industry resistance. There is still a plausible late-session surprise if negotiators attach a narrow AI provision to a must-pass package, but that would require multiple contingencies to align at once. Overall, the evidence supports a cautious estimate that enactment is possible but unlikely, with the No side remaining clearly favored.
Arguments
For
- Arguments for Yes: Bipartisan concern about AI risk is real, which increases the odds that lawmakers eventually find a narrow compromise.
- Arguments for Yes: Multiple proposals already exist, so Congress does not need to start from scratch if leadership decides to prioritize the issue.
Against
- Arguments against Yes: The House has not shown meaningful floor momentum, which is usually necessary before enactment can happen.
- Arguments against Yes: The central policy disputes are still unresolved, and the remaining time before the deadline is limited.
Key drivers
- The House calendar is tight and major AI bills have not reached the floor.
- Senate negotiators have not finalized bipartisan language for a bill.
- Election-year congestion makes new standalone legislation harder to move.
- Committee-level progress suggests some chance of a late compromise.
Risk factors
- A last-minute AI provision could be attached to a must-pass spending or defense bill.
- A bipartisan Senate breakthrough could accelerate if leadership prioritizes AI safety in a short post-election session.
Scenarios
Best case
Negotiators quickly settle on a narrow bipartisan AI safety bill, leadership finds a way to move it through both chambers, and the President signs it before the end of 2026.
Most likely
Congress continues debating AI safety but fails to complete the full legislative process in time, leaving the market to resolve No.
Worst case
Congress remains locked in procedural delay, no consensus text emerges, and all meaningful federal AI safety efforts are pushed into 2027 or shelved.
More from this day
- PoliticsKalshiEnded
When will a reconciliation bill become law?
AI99%MKT2%Edge+97Hidden GemA reconciliation bill appears to have already become law, which would satisfy the condition well before Oct. 1, 2026. On the facts provided, I put the Yes probability extremely high unless the market uses an unusually narrow resolution rule tied to a different bill.
- PoliticsKalshiEnded
When will Trump nominate a Federal Reserve governor?
AI99%MKT3%Edge+96Hidden GemA nomination has already been made, so the Yes outcome is overwhelmingly likely unless the market is using an unusually narrow definition of nomination. The current price looks like a clear misread of the event timing and the underlying news flow.
- PoliticsKalshiEnded
Will 3 mg THC beverages qualify as hemp under federal law?
AI9%MKT87%Edge-78HypedA 3 mg THC beverage is very unlikely to qualify as hemp under the federal changes taking effect in November 2026. My independent probability for a Yes outcome is 9%, far below the market price.