EVwire brief: Waymo says its driverless vehicles have covered more than 220 million fully autonomous miles through March 2026, and over that distance, the Waymo Driver was in 94% fewer serious-injury-or-worse crashes than human drivers would have been on the same roads.
Measured against the human benchmark in its operating cities, that works out to 45 fewer serious-injury-or-worse crashes, 306 fewer airbag-deployment crashes (82% fewer), and 707 fewer injury-causing crashes (also 82% fewer).
Waymo's March 2026 safety update, covering more than 220 million rider-only miles.
Waymo is now driving more than 4 million miles a week, which it says translates to roughly one fewer serious-injury-or-worse crash every eight days, plus six fewer airbag-deployment crashes and 13 fewer injury-causing crashes every week.
“Despite expanding into increasingly complex environments ike airports and new cities, the Waymo Driver’s safety performance remains exceptionally robust.”
Here's Waymo's update on LinkedIn:
The comparison is the heart of it. Waymo stacks its rider-only crash rate against how often human drivers crash on the same surface streets in the cities where it runs, Phoenix, San Francisco, Los Angeles, and Austin, using crash reports filed under NHTSA's Standing General Order.
What Waymo is selling in this update is consistency. The company says that even as it has pushed into harder settings like airport runs and brand-new cities, the Waymo Driver's numbers have held, rather than slipping as the driving got more complex.
Waymo’s fleet now drives 4 million miles per week.
Context:
It is worth being precise about what the figures are and are not. They compare Waymo to the overall human driving population in those cities, on surface streets rather than freeways, and Waymo logs even minor contact while a large share of human crashes never reach a police report. Waymo adjusts for that underreporting on the injury benchmark, but it is a modeled comparison, not a direct head-to-head.
Here is where it earns trust, though. The methodology is peer-reviewed and built on published best practices. Outside groups like the Insurance Institute for Highway Safety (IIHS), the University of Michigan, and Swiss Re have vouched for the approach, and the raw crash data sits on Waymo's site as downloadable files so anyone can rerun the math. A separate Swiss Re insurance study found similar drops in liability claims as Waymo expanded into new cities.
Source: Waymo's Safety Impact hub and its safety update on LinkedIn
DON'T FORGET to subscribe for the day's EV and AV news, and join 14,000+ EV geeks.
