The big picture: As Tesla's stock climbs to record highs on hopes of a trillion-dollar driverless taxi market, its early robotaxi rollout in Austin, Texas, illustrates the gap between investor expectations and on-the-ground reality. The company has deployed about 30 robotaxis in the city since launching service in June, compared to roughly 200 vehicles from Waymo, Alphabet's autonomous driving division, which began operating in Austin in March. Waymo already runs paid rides in four other cities and oversees a fleet of more than 2,500 vehicles, while Tesla's robotaxis in Austin and San Francisco still rely on human monitors sitting in the cars.
Austin, better known for its music venues and startup scene, has quickly turned into what city transportation official Lewis Leff calls a "petri dish" of autonomous technology.
In Austin's mix of vehicles, Waymo uses a Jaguar I-Pace SUV, Tesla relies on its Model Y, and Zoox operates a steering-wheel-free, rectangular vehicle with sliding doors. Zoox, Amazon's driverless car unit, tested its custom-built shuttles there before expanding commercial service to Las Vegas and San Francisco. Uber has also moved into the region by testing a driverless service in Dallas through Avride, an Austin-based company, while Aurora and other firms experiment with self-driving trucks elsewhere in Texas.
Waymo's stronger foothold reflects a much earlier start. Google's self-driving car project began in 2009 and later became Waymo, which launched its first commercial robotaxi service in Phoenix in 2018. Tesla entered the race later but with aggressive promises from chief executive Elon Musk, who said Tesla cars would drive themselves across the United States by 2018 and predicted one million robotaxis on the road by mid-2020. Those timelines have slipped, even as Tesla has continued to pitch autonomous driving as central to its future.
The two companies also diverge sharply on hardware. Tesla's system relies solely on cameras to perceive its surroundings, a choice aimed at lowering costs while competitors use more expensive sensor suites.
Waymo and Zoox combine cameras with radar and lidar to achieve greater redundancy and range. Critics say Tesla's camera-only strategy struggles in conditions such as low visibility or heavy glare. "I'm still deeply skeptical that Tesla is all that close in terms of building a real automated driving system," Matthew Wansley, a professor at Cardozo School of Law and former autonomous driving executive, told The New York Times.
Despite those doubts, Musk has continued to frame autonomy as transformative for Tesla's business model. "We're really just at the beginning of scaling quite massively Full Self-Driving and Robotaxi and fundamentally changing the nature of transport," he told investors in October. Shareholders recently approved a compensation plan that could be worth up to $1 trillion, tied in part to the commercial deployment of one million robotaxis.
Some analysts question whether the financial payoff can match the rhetoric. HSBC analyst Michael Tyndall has argued that even reaching hundreds of billions of dollars in robotaxi revenue would require many people to give up personal car ownership, which he views as unlikely in the near term. "There is quite a lot of overhead which is not talked about," he said.
Other investors remain more patient. RBC's Tom Narayan said investors are "not that concerned as long as we're moving in the right direction," reflecting confidence that Tesla could eventually leapfrog rivals by pushing software updates to millions of cars already equipped for autonomy.
On the streets of Austin, officials and residents are getting an early look at how those competing strategies play out. City officials say both Tesla and Waymo vehicles have been involved in minor incidents, though they emphasize the autonomous cars are not always at fault. Waymo recalled vehicle software after reports that some of its cars had driven past stopped school buses and said it has since improved its systems.
For Tesla, early ride data in Austin suggests its robotaxis are competent but still rough around the edges. Some cars have monitors in the driver's seat; others have them in the front passenger seat, but in each case, a human is present.
Several trips proceeded smoothly through city traffic, yet the vehicles occasionally made questionable decisions, such as entering an intersection on a late yellow light or stopping short of the intended drop-off point. Availability has also been inconsistent, although fares have been low, with a 20-minute ride across the city costing just over five dollars.
Texas's regulatory environment has helped make the state attractive to autonomous-vehicle companies, but oversight is tightening as deployments grow. Beginning next May, companies will be required to secure approval from the Department of Motor Vehicles before operating services. "It's not 'Move fast and break things and get to market.' It's deliver a quality solution that is safe for the public," said Darran Anderson of the state's transportation department.