Shares of ride-hailing giant Uber Technologies (UBER 1.16%) have rebounded recently as investors warm to the idea that robotaxis could become a major new growth driver. Shares are up about 7% over the last month and 5% in the last week alone. At close to $76 as of this writing, though, the stock still sits about 25% below its 52-week high near $102. The recent optimism toward the stock is easy to understand. What's harder to pin down, however, is what Uber actually owns in the autonomous race. Here is the part the robotaxi excitement tends to gloss over: Uber doesn't build the cars, doesn't write the self-driving software, and doesn't own the vehicles carrying its riders. Its plan is to be the app that books the trip, whoever's autonomous car shows up. That asset-light approach could be Uber's biggest advantage in autonomy -- or its biggest vulnerability, depending on how the next few years unfold. A platform, not a fleet Uber's pitch to investors and autonomous-car manufacturers is about aggregation. It has reportedly signed up about 30 autonomous partners -- robotaxi developers, delivery-bot makers, and self-driving trucking firms -- and wants to be the marketplace where that capacity meets demand. The early traction backs up the idea: Uber recently said autonomous trips on its platform grew about tenfold over the past year, and management is targeting driverless service in up to 15 cities by the end of 2026. "We get to work with everybody in the ecosystem," Uber CEO Dara Khosrowshahi told Fast Company in a June interview, pointing to a network that handles more than 40 million trips a day. The logic is that with that much demand, Uber can keep a partner's expensive cars busy in ways a single operator running its own app can't. And the