Uber will invest up to $1.25 billion in EV maker Rivian as part of a deal that could see the ride-hailing giant ultimately buy as many as 50,000 robotaxis from the California-based company.
Under the deal, made public on March 19, Uber — along with its fleet partners — has committed to buy 10,000 self-driving versions of the automaker’s R2 SUV, with robotaxi operations scheduled to start in San Francisco and Miami in 2028. This initial investment stretches to $300 million.
Should a number of defined “autonomous milestones” be reached, further investment will follow as part of a plan to roll out robotaxis across 25 cities in the U.S, Canada and Europe by 2031. Uber and its fleet partners will also have the option of buying another 40,000 autonomous robotaxis in 2030.
Over the past 18 months, Uber has been busy signing a series of deals with automakers, AI vendors and tech giants as part of its long-term vision to offer self-driving options across its fleet globally.
Among the more prominent collaborations are with Waymo in Phoenix, Austin and Atlanta, plus in Las Vegas with Motional, where it is also planning to work with Amazon’s Zoox.
Further afield, Uber has paired up with China’s WeRide in the Middle East and has revealed plans to operate its first robotaxis in Japan in tandem with British self-driving AI vendor Wayve and automaker Nissan.
“We’re big believers in Rivian’s approach – designing the vehicle, compute platform, and software stack together, while maintaining end-to-end control of scaled manufacturing and supply in the U.S.,” Uber CEO Dara Khosrowshahi said in a statement.
“That vertical integration, combined with data from their growing consumer vehicle base and experience managing the complexities of commercial fleets, gives us conviction to set these ambitious but achievable targets,” he said.
The R2 features Rivian’s third-generation autonomy platform, which includes a multimodal sensor suite comprising 11 cameras, five radars and one Lidar. The platform is driven by two of Rivian’s in-house RAP1 AI chips.
The platform uses a continuously self-improving data flywheel that learns about human driving using real-world data from the customer fleet, with the knowledge base expanding in real time.
This base fuels Rivian’s Large Driving Model. According to Rivian: “Its LLM-like architecture enables us to apply advances in generative AI directly to our autonomy stack, including reinforcement learning, which distills superior driving strategies to the on-board models with minimal compute overhead.”
The Uber deal could serve as vindication for Rivian CEO CJ Scaringe, who made clear how pivotal automation was to the company’s future at its first-ever Autonomy and AI Day in December.
