Waymo is way, way ahead on testing miles–that might not be a good thing
arstechnica.comOh.. arstechnica always finding a way not to talk about Tesla.
The author explained in a comment why Tesla wasn't included. tl;dr They reported 0 miles and agree that they shouldn't be included.
> "I see Tesla as being in a separate category. Right now Tesla's autonomy stack has much more limited functionality than Waymo's (or Cruise, Argo, Zoox, etc.). And it's overseen by ordinary customers, not professional safety drivers. That means that Tesla doesn't get the kind of annotated feedback Waymo gets. If a customer deactivates Autopilot that _might_ mean that it made a mistake, but it also might just mean that the driver decided to pull over to buy a cup of coffee. Tesla agrees with me—they reported 0 miles of on-road testing in California in 2018.
"I think you can see Tesla as pursuing a third strategy alongside the Waymo/Cruise taxi strategy and the Voyage/Nuro 'low-speed services first' strategy. I'm personally not optimistic about Tesla's strategy but I can't rule it out."
This is like some expert explaining how Apple's approach to smartphones is not good as Nokia's. The tech is in its infancy. There are no experts other than the players themselves.