Growing proof that autonomous cars save lives
- Plagued by skepticism, AVs still show reduced crash rates and fewer injuries compared to human drivers, with AEB systems cutting pedestrian crashes by 27% and rear-end crashes by 50%. These systems are mandated for all light vehicles in the U.S. by 2029, and over 90% of new models already comply. The Highway Loss Data Institute found that cars with AEB alone reduced property-damage claims by 13%, while bundling ADAS features reduced claims by up to 39%.
- Waymo’s robotaxis have given 20 million paid rides over 220 million miles, equivalent to 250 lifetimes of driving, and have recorded 92% fewer fatal or serious-injury crashes, a 13-fold reduction versus human drivers in comparable city environments. They also have 92% fewer pedestrian injuries, 83% fewer airbag deployments, and 82% fewer injuries. These figures are based on limited, fair-weather data in a few cities.
- The IIHS is concerned about the comparability of Waymo’s data to human drivers, citing that up to one-third of human crashes go unreported and that insurance price hikes may skew safety numbers in favor of humans. Despite this, Waymo’s injury crashes were 81% lower on a per-mile basis. The IIHS calls for better federal reporting standards to facilitate meaningful comparisons.
- Amazon’s Zoox receives an exemption from safety standards, allowing commercial operation of its robotaxis, which have no steering wheel or pedals. NHTSA also partners with SAE Industry Technologies to develop AV standards. The IIHS emphasizes the need for national performance and safety standards to enable meaningful crash rate comparisons.
- The article highlights the potential of AVs to reduce road fatalities, with estimates suggesting that a 50% reduction in road deaths could save 580,000 lives annually. It underscores the importance of public trust and the need for standardized safety reporting to support AV adoption.
AI-generated summary of the original article — see the source for the full story.
Log in to join the discussion.