Listen to this episode on Apple Podcasts, YouTube, Spotify, and Substack. Transcripts are also available by clicking on the “transcripts” button in the top right corner of this post.
In today’s episode, I’m speaking with Dr. Missy Cummings, professor and director of the Mason Autonomy and Robotics Center at George Mason University, and former Senior Safety Advisor at the National Highway Traffic Safety Administration (NHTSA). We go deep on one of the most misunderstood pieces of the AV stack: remote operations. Dr. Cummings breaks down the difference between remote assistance and true teleoperation, why latency matters more than most people realize, and why “median latency” can be a misleading metric when safety is on the line.
We also unpack common misconceptions around Waymo’s remote assistance model, where it works well, and where it could improve. Dr. Cummings shares her perspective on whether offshore remote operators pose safety risks, and why local licensing, training, and regulatory clarity are essential as the industry scales.
Our discussion then turns to the lessons from Cruise’s collapse, what remote operations revealed about the limits of current AV systems, and how regulators should think about oversight without stifling innovation. We examine Waymo’s remote operator-to-vehicle ratio, and wrap up with a broader question: will AVs ever be fully autonomous? And what should policymakers, investors, and the public actually look at when evaluating safety claims and remote operations?
Chapters
(00:00) Introduction to Dr. Missy Cummings
(05:55) Remote assistance vs remote driving (teleoperation): characteristics and differences
(10:45) Effects of time latency on remote operation
(14:55) Why “median latency” is misleading and why maximum latency is the number that really matters.
(17:02) The misconceptions around Waymo’s remote assistance process, and how it could improve
(20:26) Are offshore remote operators a safety risk? The importance of local licensing and training
(26:06) Remote operation lessons from Cruise’s demise
(27:05) What should be the limits of regulatory oversight in autonomy?
(33:14) Waymo’s remote operator-to-vehicle ratio, and the rarity of fully autonomous systems.
(37:39) Will AVs ever be fully autonomous?
(39:25) What the public and regulators should look out for when evaluating Waymo’s safety record and remote operations.
(42:05) Conclusions and final thoughts
Notes/Links:
You can find Dr. Cummings on LinkedIn, where she’ll announce the publication of her upcoming article mentioned at the 06:35 timestamp.
Brad Templeton’s article I mentioned at the 12:13 timestamp: Waymo Runs A Red Light And The Difference Between Humans And Robots (link, no paywall).
Waymo’s latest blog post on their remote assistance, mentioned at the 14:52 timestamp (link).
Waymo’s earlier blog post outlining the “Phone a friend” structure of their fleet response I mentioned at the 17:04 timestamp (link).
John Deniston’s TDD guest post I mentioned at the 24:41 timestamp (link).
-Harry
