The SELF DRIVE Act of 2026 establishes federal safety requirements, testing rules, and data reporting standards for the safe deployment and research of automated driving systems in vehicles.
Robert Latta
Representative
OH-5
The SELF DRIVE Act of 2026 establishes a federal framework for the safe testing and deployment of automated driving systems (ADS) across the United States. It mandates that manufacturers submit a comprehensive "safety case" demonstrating their ADS will not pose an unreasonable risk before vehicles can be sold. Furthermore, the bill creates a national repository for automated vehicle crash data and clarifies federal preemption over state laws regarding ADS sales and reporting.
The SELF DRIVE Act of 2026 is the federal government’s attempt to finally put a roadmap together for self-driving cars. Instead of a patchwork of state rules, this bill creates a unified national framework for how these vehicles are built, tested, and monitored. By September 30, 2027, manufacturers will have to prove their tech is safe through a 'safety case'—a detailed evidence file showing the car can handle everything from heavy rain to avoiding a cyclist. It also sets up a National Automated Vehicle Safety Data Repository to track crashes within 30 days of them happening, moving away from the current system of occasional reports to a centralized, public database.
Under Section 3, carmakers can’t just put a robot taxi on the street and hope for the best. They have to submit a safety case that explains exactly what the car can and cannot do—what the bill calls its 'Operational Design Domain.' For you, this means if a car is only rated for sunny suburban streets, it shouldn't be trying to navigate a blizzard in a mountain pass. The bill also mandates that Level 4 and 5 vehicles (the ones that don't need you to pay attention) must have a way for a passenger to command the car to reach a 'minimal risk condition'—essentially a panic button that safely pulls the car over if something goes wrong. This is a big deal for peace of mind, but the bill’s language on what counts as an 'unreasonable risk' is still a bit fuzzy, leaving a lot of power in the hands of the Secretary of Transportation to decide what’s 'safe enough.'
Section 4 opens the door for companies to test these vehicles in the real world, and not just for research. The Secretary can allow 'limited commercial operations,' which means you might see self-driving trucks hauling freight or autonomous vans delivering your groceries as part of a pilot program. To speed this up, the bill allows these test vehicles to bypass some traditional safety standards (like requiring a steering wheel or brake pedals) as long as they are being used for evaluation. While this helps tech move faster, it also means there could be vehicles on your local roads that don't meet the same crash-test or manual-control standards as the car you drive to work.
One of the most significant shifts in this bill is 'Federal Preemption.' Essentially, Section 3 tells states they can’t pass their own laws to block the sale or manufacture of these cars if they meet federal standards. While your state still handles things like registration, insurance, and traffic tickets, they lose the ability to set their own unique safety or crash-reporting rules. For a delivery driver or a parent, this means the rules for the autonomous car next to you will be the same whether you're in Ohio or Oregon. However, some might worry that this 'one size fits all' approach could prevent cities from acting quickly if they notice a specific safety issue on their unique streets.