Waymo has been tracking its safety statistics closely, of course, and even sharing them with other researchers and the public to allow for independent evaluation of how well it’s doing. The short story is: Waymo robotaxis are much safer than human drivers. The company has updated us on some of the top-line stats in a post today. They are as impressive as ever. “Our latest analysis of the Waymo Driver’s safety performance now covers more than 220 million fully autonomous miles through the end of March 2026 — the equivalent of over 250 human lifetimes behind the wheel. We’ve analyzed five of our operating geographies, including Atlanta for the first time,” Waymo shares. “Our safety performance remains strong. Compared to human drivers in the same areas over the same period, the Waymo Driver was involved in 94% fewer crashes causing serious or fatal injuries, 82% fewer crashes in which an airbag deployed, and 82% fewer crashes involving any reported injury. Each comparison is made regardless of who was at fault in a collision. “This strong safety performance extends to the most vulnerable people on the road: Waymo experienced 93% fewer injury-causing crashes involving pedestrians, 84% fewer involving cyclists, and 84% fewer involving motorcyclists.” Perhaps some of these stats are a bit generous. If Waymo only compared itself to new cars with modern ADAS features, perhaps the numbers wouldn’t look so amazing. However, one can’t deny that these figures look good when comparing to the average human driver on the road in these cities. And at least the company is comparing only in the urban geographies where its robotaxis are in operation. Also, the company notes that its improved safety stats have held steady even as the company has significantly expanded service. “What stands out in this latest update is consistency of safety performance. Since our last analysis, we unified our San Francisco Bay Area service area, expanded into more complex environments including airport service, and reached statistical significance in Atlanta, with its own road layouts and traffic patterns. The operating environment changed substantially, but the safety performance remained robust.” Waymo also chooses to highlight this in its newest safety-stat area, Atlanta. “Atlanta is a clear illustration. Across more than 5.4 million fully autonomous miles there, the Waymo Driver was involved in 94% fewer airbag-deployment crashes and 86% fewer injury-involving crashes than the human benchmark — both statistically significant. A typical driver covering those same miles in Atlanta would have been expected to be involved in roughly 1.2 crashes causing serious or fatal injuries. The Waymo Driver was involved in none.” The company also does its best to humanize the data and impress upon all of us how much this matters. “These percentages describe a rate, but what they add up to is people. Waymo now drives more than 4 million miles every week. That translates to an estimated one fewer serious-injury-or-worse crash every eight days, and roughly six fewer airbag-deployment crashes and thirteen fewer crashes that cause any kinds of injuries each week. “Over the lifetime of our operations, these safety benefits accumulate for our riders and the communities we serve. Waymo has experienced an estimated 47 fewer crashes causing serious or fatal injuries, 305 fewer crashes that led to the deployment of an airbag in any involved vehicle, and 707 fear crashes causing injuries of any kinds than would be expected had those miles been driven by humans in the same places. As the Waymo Driver drives more miles, those totals keep growing.” It’s hard to argue with that. And it’s hard to not say Waymo robotaxis should be deployed everywhere. If you want to check out the data and methodology for yourself, you can do so here. To wrap up, University of Michigan Transportation Research Institute Professor Carol Flannagan says: “Waymo has reached the point where it is driving enough miles to make direct comparisons to human drivers on crash rates. The consistency of the results across locations helps support the strength of the conclusions, and their methods give significant attention to the challenges of making apples-to-apples comparisons when the data sources for humans and the Waymo vehicle are different. The focused results at intersections are particularly positive because intersections remain a significant safety challenge, especially for vulnerable road users.” Again, I do fully celebrate these improvements. However, it would be nice to see even more refined research that compared Waymo robotaxis to vehicles of the same age with common ADAS features. That would be an ideal comparison that would really show us how much Waymo Driver improves safety on the road.