Tesla unveiled a purpose-built robotaxi vehicle and outlined plans for a ride-hailing network, according to Reuters, doubling down on a vision-only approach to autonomous driving in contrast to the sensor-heavy architecture used by Waymo.
The bet is the clearest expression of a long-running argument in the autonomous industry: whether a sufficiently capable software stack, trained on the vast driving data Tesla collects, can replace the lidar and redundant sensors that the companies operating driverless services today rely on. The two approaches have not yet been tested against each other at commercial scale.
What the reveal did and did not show
The reveal showed a vehicle and a timeline; it did not show a service operating without a safety driver in the way Waymo's already does. The market's response reflected that distinction. The robotaxi business, if it arrives, will be judged on the same metric every autonomous service is judged on: paid rides, in a city, without a human in the driver's seat. Tesla has the distribution and the data; it has not yet shown the service.

