741,195 Fords Can’t Stay in Park. NHTSA Just Proposed Removing Brake Pedals.

Two things happened in the same seven-day window at the National Highway Traffic Safety Administration. On June 24, Ford filed recall 26V402, covering 741,195 Expeditions, Navigators, Explorers, Aviators, and F-150s whose transmissions can damage their own parking systems, leaving the vehicle unable to hold in Park.^[1] On June 25, NHTSA published a proposed rule to amend Federal Motor Vehicle Safety Standard No. 135, removing the requirement for a brake pedal in vehicles operated entirely by automated driving systems.^[2]

Read that sequence again. Day one: three-quarters of a million trucks and SUVs can roll away while their owners walk into Costco. Day two: let’s take the brake pedal out of the next generation of cars entirely.

The recalled Ford models span the 2018 through 2021 model years, meaning most have been on the road five to eight years. At the national average of roughly 12,000 miles per year, that fleet has accumulated an estimated 49 billion miles with a transmission defect that can prevent the parking system from holding. Waymo, the largest autonomous vehicle operator in the country, announced 200 million fully autonomous miles as of June 2026, the result of more than a decade of development, 3,871 robotaxis, and operations across ten metro areas.^[3] Ford’s defective fleet has driven roughly 245 times further than every Waymo robotaxi combined, across the company’s entire existence.

NHTSA Administrator Jonathan Morrison called the brake-pedal proposal part of an effort to tear down “pointless barriers to innovative designs.”^[4] The proposed rule would not change stopping-distance requirements, only eliminate the physical pedal in AV-only vehicles. What it does not address is how a passenger would stop the vehicle in an emergency if the software fails. The Drive described this as a question NHTSA is “leaving for another day.”^[2]

Ford, meanwhile, has filed 51 recalls since January 1, 2026—more than any other automaker and nearly three times the runner-up, Chrysler, at 19.^[5] The recall surge traces directly to a 2024 consent order with NHTSA that installed independent oversight of Ford’s safety process. Michael Brooks, executive director of the Center for Auto Safety, told USA Today that “Ford’s recall numbers started creeping up to significant levels right about the time the consent order was in the works.” Ford’s response: more than 80% of its 2026 recalls are “resolved through convenient, software-only updates.” This particular one is not. The transmission fix will not be available until April 2027, a full ten months after the recall was filed.^[1]

In the FARS database, the recalled models carry real weight. The F-150 has recorded 9,194 fatalities over the 2014–2023 period at a rate of 1.04 deaths per 100 million vehicle miles traveled. The Explorer sits at 3,797 deaths (1.54 rate). The Expedition: 1,515 deaths at 2.31, more than double the F-150’s per-mile risk, and the eighth-deadliest SUV in the dataset. Ford as a brand accounts for 34,954 FARS fatalities across 29 models, second only to Chevrolet.^[6] These are not boutique products with niche safety profiles. They are the most common vehicles on American roads.

The juxtaposition is not about whether autonomous vehicles should eventually operate without brake pedals. It probably makes engineering sense for a purpose-built robotaxi to eliminate controls no one will use. The question is sequencing. NHTSA is fast-tracking the deregulation of a technology that has logged 200 million miles total while processing recalls for conventional vehicles that have accumulated 49 billion defective miles and won’t see a repair for almost a year. The agency with 550 employees and a $1.2 billion budget is simultaneously supposed to oversee 290 million registered vehicles and build the regulatory framework for an entirely new category of machine. Something will not get adequate attention, and historically, the thing that loses is the boring recall follow-through.

The strongest counterargument: these are different risk domains. A rollaway defect has a mitigation—Ford says to use the parking brake every time, which you should do anyway. A brakeless robotaxi operates in a fundamentally different failure mode where software replaces the entire braking chain. Comparing them directly is, in some respects, like comparing a broken deadbolt to a missing steering wheel. Fair enough. But the regulator processing both issues is the same understaffed agency, working the same week, with the same finite attention. And one of those issues affects a fleet 245 times larger than the other.

What to do: If you own a 2018–2021 Expedition or Navigator, a 2020–2021 Explorer or Aviator, or a 2021 F-150, check your VIN at nhtsa.gov/recalls. Use the parking brake every single time you park. Do not wait for the notification letter, which Ford expects to mail between August 3 and August 9. And do not assume the fix is imminent: the software update remedy is currently scheduled for April 2027.

Limitations: The 49-billion-mile estimate assumes all 741,195 vehicles carried the defect from manufacture; actual onset may vary. VMT per vehicle uses the national average of ~12,000 miles/year; trucks and SUVs typically run higher, meaning the true figure could be larger. Waymo’s 200 million miles count only fully autonomous driving; total industry AV miles including safety-driver testing are higher but still orders of magnitude below the defective Ford fleet. FARS data covers 2014–2023 and reflects all model years in the fleet, not just the recalled cohort.