Aisha works the overnight shift at a hospital in Phoenix. Most nights, she bikes home at 4 a.m. because the bus doesn’t run that early and a cab eats into money she’d rather save. She told me once that the scariest part of her whole shift isn’t the ER, it’s the ten-minute ride home past bar traffic on a dark street. That’s the exact kind of trip Waymo’s new safety data is talking about.
On September 24, Waymo released fresh numbers covering 270 million fully driverless miles across five cities: Los Angeles, San Francisco, Phoenix, Austin, and Atlanta. The company says its robotaxis prevented at least 841 injury-causing crashes, compared to what an average human driver would have caused over those same miles. That’s not a slogan. It’s Waymo’s own published safety data.
What Waymo’s Safety Data Actually Means for You
Here’s the part that matters if you walk, bike, or ride a motorcycle anywhere near one of these cities. The biggest safety gains aren’t for the passengers inside the car. They’re for the people outside it.
- 82% fewer crashes that cause any injury, compared to human drivers covering the same distance
- 95% fewer crashes serious enough to send someone to the hospital
- 93% fewer pedestrian injury crashes
- 86% fewer cyclist injury crashes
- 82% fewer motorcyclist injury crashes
Think about Aisha’s ride home. A tired or distracted driver rolling through a crosswalk is exactly the scenario these numbers describe. A car that never gets drowsy, never checks a text, and reacts in a fraction of a second changes the odds for her every single night.
The Money and Time This Could Save You
Injury crashes are expensive for everyone, not just the person hurt. A moderate-injury crash typically costs well past $100,000 once you count medical bills, lost work, and vehicle damage, based on federal crash-cost estimates. Multiply that by 841 avoided crashes and you get tens of millions of dollars that never left anyone’s pocket or raised anyone’s insurance premium. It also means fewer mornings where your commute turns into a parking lot because of a wreck three blocks ahead.
Why I Trust This Data (With a Caveat)
I stay skeptical of safety claims from tech companies by default. What sets this release apart is that Waymo published the underlying crash and mileage data, not just a press release. David Zuby, chief research officer at the Insurance Institute for Highway Safety, put it well:
“By making detailed information about crashes and miles driven publicly accessible, Waymo’s transparency will not only support independent research but foster public trust,” Zuby said, urging other autonomous vehicle developers to follow suit.
The IIHS has reviewed Waymo’s crash record independently before and reached similar conclusions, which carries more weight with me than any company’s own report. Nashville wasn’t part of this round, so it’s worth watching what that market shows later.
None of this makes robotaxis flawless. They still get stuck sometimes, and they can’t help anyone in a city without the service, which is most of the country. Deployment questions remain, like the ones our recent look at NHTSA’s new safety alliance raises, and weather is still a real test, as our story on Waymo’s flood safety fix in San Antonio showed.
Still, if you’re a parent walking a stroller across a crosswalk, a delivery cyclist working a Friday night shift, or someone like Aisha pedaling home after a 12-hour shift, this data points in one direction: safer streets, measured in real avoided injuries, not projections. Ask your city whether it’s tracking these numbers for the roads you actually use, because that’s the question that decides whether this progress shows up where you live.