Urgent.News

What's breaking now, across thousands of outlets.

Tech

TechCrunch Mobility: The hidden human cost of robotaxis

Welcome back to TechCrunch Mobility, your hub for the future of transportation and now, more than ever, the role AI is playing in it.

In the world of self-driving vehicles, there is a hidden cost that comes with the advancement of technology. Sean O'Kane, a senior reporter at TechCrunch, investigated data submitted to the Occupational Safety and Health Administration and discovered that Waymo and Zoox test drivers experienced over two dozen injuries in 2024 and 2025.

These injuries were primarily caused by hard braking or sudden movements made by the autonomous vehicles, resulting in conditions such as whiplash. Some workers were sidelined for months following these incidents.

While it is tempting to assume that other autonomous vehicle developers are not experiencing similar issues, it is probable that other test drivers are also facing these challenges. However, these companies may be exempt from OSHA's reporting requirements. For further information on this story or other topics, you can reach out to Kirsten Korosec at kirsten.korosec@techcrunch.com or Sean O'Kane at sean.okane@techcrunch.com.

Written by urgent.news from TechCrunch's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at techcrunch.com →

More in Tech

What should the next Weekend Challenge theme be??

The next DEV Weekend Challenge launches next week, and we have some ideas up our sleeves, but we thought crowd-sourcing the theme would be a fun way to collect ideas from the community.

  • Next DEV Weekend Challenge theme to be determined by community input
  • Theme should be globally-minded, inclusive, and technology-agnostic
  • Theme revealed on September 3rd at 2:00 AM UTC

What stops your agent shipping garbage?

The tests, in principle. But the agent wrote those too, out of the same understanding of the task, so they agree with the code even when the code is wrong.

  • Agents can break their own tests to evaluate effectiveness
  • Placebo proofs in CLAUDE.md file identify flawed tests
  • Fixing weakest checks first ensures robust testing

More from Sunday 30 August →