Waymo's autonomous vehicles were involved in 92% fewer crashes causing serious or fatal injuries compared to human drivers, according to Waymo. A 92% reduction in crashes causing serious or fatal injuries represents a significant improvement in road safety, potentially mitigating severe trauma from traffic incidents. The quantifiable safety advantage of Waymo's robotaxis now sets a new benchmark for transportation performance.

Despite this proven safety record, human drivers remain the default for most road transportation. This societal reliance on human operation exists even as new data and sophisticated models highlight a significant safety deficit compared to Waymo's robotaxis.

The combination of Waymo's superior safety and a deeper scientific understanding of human driving risks will accelerate the push for broader autonomous vehicle adoption. This development could reshape future transportation safety standards.

  • The new computational model developed by TU Delft provides independent scientific validation for Waymo's real-world safety statistics, moving beyond mere correlation to a deeper understanding of why autonomous vehicles are safer.
  • The Reference Driver model predicts human behavior leading up to a crash, revealing that the safety deficit of human drivers extends beyond instantaneous reactions to broader patterns of decision-making.
  • Waymo's position as the only company currently offering robotaxi rides, combined with its demonstrably superior safety record and scientific backing, suggests an accelerating lead in autonomous vehicle deployment.
  • Human drivers remain the default on roads, but new data and models reveal their significant safety deficit compared to Waymo's robotaxis.

How Does the New Driver Model Work?

Scientists at TU Delft, in collaboration with Waymo, have developed a new computational model to predict human driver responses to hazardous traffic situations, according to TU Delft. This Reference Driver model closely matched human driver decisions across various tested hazardous scenarios. It displayed realistic braking reaction times and comparable choices between braking and steering.