NHTSA Had Four Years to Update Its Safety Ratings. Automakers Said No.

In March 2022, NHTSA told automakers it planned to overhaul the five-star safety rating program. Add pedestrian automatic emergency braking and require blind-spot intervention. Make lane-keeping assist standard for any vehicle that wanted the top rating. Four years later, the automakers said they weren't ready, and the government said okay.

Let me back up. NHTSA's New Car Assessment Program (the five-star ratings you see on dealer window stickers) hasn't had a meaningful structural update in years. Crash-avoidance technologies were supposed to join the rating criteria: blind-spot warning and intervention, lane-keeping assist, and pedestrian-detecting AEB that could stop a car from 62 mph before hitting a human being, all scheduled for the 2026 model year. Blind-spot monitoring and lane-keeping are already standard or available on the majority of new vehicles sold in the U.S., including most mid-trim models. This isn't a technology problem.

Then the Alliance for Automotive Innovation, the trade group representing GM, Ford, Toyota, Volkswagen, Hyundai, Nissan, Kia, BMW, Honda, JLR, Mitsubishi, Mazda, and effectively every major automaker selling cars in the United States, filed a complaint in April 2025.^[1] Their argument: NHTSA hadn't published finalized pedestrian crash test procedures, which meant manufacturers couldn't conduct their own tests, couldn't earn ratings, and would lose stars on the window sticker. So NHTSA delayed everything to the 2027 model year.^[2]

Fair point about the test procedures. Genuinely fair. If the government tells you to pass a test but doesn't tell you what the test looks like, that's an administrative failure. But this wasn't a surprise regulatory ambush sprung on an unsuspecting industry. NHTSA floated these exact changes in March 2022.^[3] Many vehicles already met or exceeded the proposed standards when the complaint was filed. The Carscoops editorial board put it plainly: "Manufacturers were hardly caught off guard."^[4]

During those four years of advance notice, pedestrians kept dying. Between 2009 and 2023, pedestrian fatalities in the United States surged 80 percent — from 4,109 to 7,380 — while all other traffic deaths rose just 13 percent.^[5] The U.S. pedestrian death rate climbed 50 percent between 2013 and 2022. Twenty-seven other high-income countries saw their rates drop by a median of 24.7 percent over the same period.^[6] America now has the highest pedestrian death rate among every high-income country the CDC examined. Three out of four of those deaths happen at night, a quarter are hit-and-runs, and sixty-five percent occur where there is no sidewalk at all.

These features are specifically designed to address this carnage. Pedestrian AEB detects people in the road and brakes before the driver does, while blind-spot intervention prevents lane changes into pedestrians and cyclists the driver can't see. None of this qualifies as a luxury gadget; it's the kind of engineering that separates a 2026 car from a 2016 car, even if both show five stars on the sticker.

Which brings up the uncomfortable contrast: IIHS, a private, insurance-industry-funded nonprofit, also tightened its safety criteria for 2026, on time, with no delays. For the 2026 Top Safety Pick+, IIHS now requires a good rating in pedestrian front crash prevention, upgraded vehicle-to-vehicle crash avoidance testing at higher speeds, and front crash prevention as standard equipment.^[7] "This year, we're asking automakers to make excellent protection for back seat passengers the norm," said IIHS President David Harkey, and nobody asked him to wait a year.

So where does that leave you as a buyer? Right now, the five-star rating on the window sticker of a 2026 car means less than it should. It doesn't tell you whether the car has pedestrian AEB. It doesn't distinguish between a vehicle with blind-spot intervention and one without. It is, functionally, the old test under the old criteria with the same old stars. IIHS Top Safety Pick+, on the other hand, means the car has been tested against 2026-level standards, including the pedestrian and crash-avoidance technologies NHTSA punted on. If you are shopping for a new car this year, the IIHS rating is the one doing the actual work.

The Alliance for Automotive Innovation also separately asked the Trump administration to reconsider mandatory AEB rules finalized in April 2024 that require nearly all new vehicles to have advanced AEB by 2029.^[2] They want that pushed back too. The pattern is straightforward: propose technology requirements, wait for industry to lobby against the timeline, delay, repeat.

And consider: Euro NCAP already tests vehicle-to-vehicle AEB up to 80 km/h (50 mph), and vehicles sold in Europe routinely pass.^[8] NHTSA's 62-mph requirement pushes that boundary further, but the underlying technology is the same radar-camera fusion stack that European automakers have been shipping for years. The industry knows how to build these systems. They asked for more time while the government said yes.

What This Means For You

If you're buying a new car in 2026: ignore the five-star window sticker because it's running on last year's test. Check the IIHS Top Safety Pick list instead. That rating actually includes the pedestrian and crash-avoidance technologies NHTSA delayed. Make sure pedestrian AEB, blind-spot intervention, and lane-keeping assist are standard on the trim you're buying, not optional packages buried in a $3,000 technology bundle. And check your VIN at nhtsa.gov/recalls regardless.

Limitations

Five-star ratings are voluntary. Automakers aren't legally required to earn stars; the system gives consumers information, not mandates. A one-year delay to 2027 is not a cancellation. And NHTSA's failure to publish finalized test procedures was a genuine administrative shortcoming, not just an excuse. But having four years of advance notice and still not being ready is a pattern that repeats across this industry in every safety domain. The test-procedure complaint was the exit ramp, and the real objection was the timeline.

The Counterargument, at Full Strength

If NHTSA publishes a standard without publishing the test procedure, automakers genuinely cannot test their vehicles against it. They cannot submit results they don't have, and the Alliance's complaint was procedurally valid: the agency told the industry what to achieve but not how it would verify achievement. That's a regulatory failure, and blaming automakers for exploiting it ignores that the government didn't do its own job. Moreover, rushing vehicles into production to meet an incomplete standard could introduce its own safety risks: poorly calibrated pedestrian AEB systems that phantom-brake on highway overpasses, for instance, trade one danger for another, and in this reading the delay isn't obstruction but due diligence.