Browse
Log in / Register
Share
Waymo's autonomous taxis resumed highway operations in Phoenix on July 29, following a 72-day pause triggered by a software recall. The company marked the return by reiterating its central safety claim: its fleet is 94% safer than human drivers in preventing serious injuries and fatalities. Yet a separate study, released just ten days before the highway restart, raises difficult questions about that benchmark.
Researchers at the Sam Schwartz Transportation Research Program at Hunter College published findings on July 20 indicating that Waymo's rate of crashes resulting in serious injuries or death is 1.5 times higher per million miles than New York City's professional for-hire fleet—the very Uber and Lyft drivers these robotaxis are designed to replace. In the San Francisco Bay Area, Waymo's largest urban market, that rate jumps to more than three times higher.
The distinction matters for consumers making a decision today. A professional Lyft driver is roughly 14 times safer than the average American motorist, according to the study. If that is the correct comparison, Waymo's safety advantage over its actual competition is far slimmer than the 94% figure implies. On the most serious crash types, the data suggests the gap may currently run in the other direction.
Methodology Dispute Remains Open
Waymo has pushed back against the study's conclusions. Spokesperson Ethan Teicher stated, "Any meaningful safety analysis should compare apples-to-apples data, based on best practices. We caution against comparisons that extrapolate Waymo's safety performance from one place to another."
The company's 94% statistic uses NHTSA's standard methodology, comparing its crash rate against the entire U.S. driving population—including impaired, distracted, and inexperienced drivers. In contrast, the Hunter College study draws on Waymo's own NHTSA filings and compares them against data from New York City's Taxi and Limousine Commission and Department of Transportation spanning back to 2020.
The methodological disagreement is not easily resolved. Waymo has not published a direct, apples-to-apples comparison of its serious-injury rate against professional for-hire drivers using its own internal data. Until it does, the debate will remain open.
The Freeway Fix: What Was Actually Repaired
Waymo's highway suspension followed a recall involving 3,871 robotaxis. The root cause, detailed in NHTSA's recall filing, was a software conflict. Waymo's 5th generation system could detect construction zone boundaries—signs, cones, flashing lights—but a behavioral decision layer prioritized avoiding oncoming traffic over staying out of the closed zone. The vehicle drove through construction barriers at speed.
An over-the-air patch, deployed between June 16 and 29, rebalanced that priority weighting. The 13 incidents that triggered the recall—six in Phoenix and seven in the Bay Area—did not result in injuries or collisions, although one passenger described his Waymo driving through an active highway work zone before a California Highway Patrol officer began pursuit.
What the patch does not fix is emergency scene recognition. NHTSA Administrator Jonathan Morrison issued a letter on July 8 declaring that the industry's inability to respond to emergency scenes constitutes a "functional insufficiency," not an edge case. Smoke degrading LiDAR and emergency lights causing sensor artifacts represent a harder perceptual problem that a simple weighting patch does not solve.
Six Recalls in Two Years
The construction zone recall is Waymo's sixth software recall since February 2024. The pattern includes robotaxis entering flooded roads, illegal behavior around school buses in Austin, low-speed collisions with parking barriers, and a collision with a towed pickup truck in Phoenix.
Waymo defends this record as evidence its self-regulatory process works: the company identifies problems, discloses them, and deploys fixes voluntarily. However, the flood recall complicates that argument. An interim patch failed in Atlanta within two weeks, and Waymo has acknowledged it still lacks a permanent fix for rain-related navigation failures.
Two federal investigations remain open: one by NHTSA and another by the National Transportation Safety Board into a January 2026 collision in which a Waymo robotaxi struck a child near an elementary school in Santa Monica. The Austin school bus recall, covering at least 24 illegal passes, is also under federal review.
Practical Guidance for Riders
For anyone deciding whether to use a Waymo for an airport trip today: In Phoenix, highway routing to Sky Harbor International Airport resumed as of July 29. The construction zone fix has been validated and has produced no documented incidents since deployment. In Los Angeles, San Francisco, Miami, and Austin, highway routing is not yet restored. Airport trips requiring highways are currently rerouted to surface streets.
The Hunter College study does not change the fact that Waymo's 220.6 million rider-only miles show a lower overall injury-crash rate than the general population. But for riders and regulators, the critical question is whether the company's safety claims hold up against the professional drivers it actually competes with—and that question now has public data to evaluate.
Latest Developments
As of today, NHTSA has not announced any new action on the Hunter College study's findings. Waymo continues to operate in its approved markets while federal investigations into the Santa Monica collision and Austin school bus incidents remain active. The company has stated it is cooperating fully with all regulatory inquiries.
Consumers should weigh the available data carefully. While Waymo's safety record is substantial, the most severe crash category shows a pattern that warrants attention—especially for those considering the service in cities where highway operations have yet to resume.
