Waymo, the undisputed heavyweight champion of robotaxis, currently operates over 3,000 driverless cars across at least 10 U.S. cities. A pack of eager also-rans - including Tesla, Zoox, Avride, and Motional - is frantically trying to catch up to the Alphabet-owned juggernaut. But what if being No. 2 isn't just a participation trophy, but a genuine strategic masterstroke?
Nuro, the delivery-robot-turned-robotaxi-aspirant founded by veterans of Google's self-driving car project, certainly thinks so. After pivoting from hauling groceries to hauling humans in 2024, Nuro struck a deal with Uber and Lucid to deploy tens of thousands of robotaxis across the US, netting itself hundreds of millions from Uber in the process. The company plans to launch in San Francisco later this year, and earlier this month, it snagged the first of several necessary permits.
Dave Ferguson, Nuro's cofounder and co-CEO, argues that Waymo's early successes - and, more importantly, its stumbles - provide invaluable free lessons. "There is a lot of value in this sort of classic second mover perspective," Ferguson said. "In some of the rare cases where they’re having challenges, [Nuro is] using those to kick the tires on our system." Ferguson speaks from experience: he got his start at Google's self-driving project alongside cofounder Jiajun Zhu, and the two left in 2016 to found Nuro. Their technology, Ferguson insists, is easily transferable from delivery to passengers - even if its passenger experience is currently zero.
Nuro's "second mover" theory means skipping the painful trial-and-error phase Waymo endured. Ferguson wants the initial robotaxi service to be broadly useful from day one, avoiding an ultra-incremental rollout. "This is not going to be only protected intersections, and then slowly we add unprotected," he said. "It’s going to be a very broad [operational design domain] to begin with." However, he noted that the launch won't cover "the entire South Bay on day one."
The Uber-Lucid-Nuro partnership is a three-headed beast: Nuro develops the sensing and compute stack, Lucid integrates it directly into the Gravity SUV on the production line (leaving the factory Level 4-ready), and Uber owns and operates the fleet, managing depots and remote assistance. Remote assistance, which has drawn congressional scrutiny, is often misunderstood, Ferguson said. "The view that the public probably jumps to when they’re told remote assistance of self-driving vehicles is someone in a dark room driving a car around like they’re playing a video game," he said. "I think that is pretty far from how remote assistance typically works." In reality, remote operators answer questions and provide prompts when vehicles get confused.
Nuro's long-term goal is to build the most capable AI driving system possible, applying lessons from both its older rules-based machine learning and newer end-to-end learning models. Ferguson acknowledged that robotaxis suffer from a lack of public trust, especially around edge cases and traffic-blocking incidents. He vowed to follow Waymo's model of transparency, sharing driving statistics to build customer confidence. "The more evidence we have of Nuro and Uber and Lucid providing a product that is dramatically safer and better for our streets than a human-driven vehicle … the better that is for everyone," he said. The company is still figuring out "the right balance of how much detail do we provide," but Ferguson is confident they'll get there.