EL7.AI
Mergers & AcquisitionsMediumUpdated×2•Originally published 22 October 2026•Updated 22 October 2026•
4 min read

Rivian received $300 million from Uber under milestone-based autonomous driving deal

Key Facts

1Rivian received $300 million from Uber in May 2026 in exchange for about 20 million Class A shares.
2Uber's targeted investment totals up to $1.25 billion through 2031, subject to achieving autonomous driving performance milestones.
3The subscription agreement provides for up to an additional $950 million across 4 remaining milestones, subject to specific conditions.
4Uber or its partners expect to purchase 10,000 robotaxis, with an option to buy up to 40,000 additional vehicles in 2030.
5The plan targets a launch in San Francisco and Miami in 2028 and expansion to 25 cities by 2031.

Rivian received $300 million from Uber in May 2026 in exchange for about 20 million shares of its Class A common stock, according to the company's filing with the U.S. Securities and Exchange Commission. The investment was part of a partnership to develop and deploy robotaxis that Uber announced on March 19, 2026. Uber at the time set its targeted investment in Rivian at up to $1.25 billion through 2031. The partnership announcement tied the investment to specific autonomous driving performance milestones. The funding Rivian received in exchange for issuing shares to Uber thus came under an agreement linking subsequent payments to progress in developing the technology. That condition puts autonomous driving system performance at the heart of the companies' funding relationship, alongside plans to purchase and operate vehicles.

The subscription agreement provides for up to an additional $950 million in total funding across 4 remaining milestones, according to Rivian's filing. These payments are subject to the agreement's terms and the achievement of its specified milestones, making completion of the funding contingent on meeting them. The company's filing described the commitment as "up to an aggregate of $950 million across the four remaining milestones," subject to certain conditions. On profitability, Reuters reported in its coverage of the partnership announcement that Rivian no longer expected to achieve an adjusted core profit in 2027. That marked a retreat from the profitability timeline the company had previously forecast. During the trading session on the day the partnership was announced, the agency said Rivian shares were up about 1% in afternoon trading.

Uber or its fleet operating partners expect to purchase 10,000 fully autonomous R2 robotaxis, according to the partnership announcement. The agreement includes an option to buy up to 40,000 additional vehicles in 2030, alongside the expected initial purchases. On the technology side, Rivian says the knowledge accumulated from driving data feeds what it calls its "Large Driving Model." The company describes the system in its official Autonomy and AI Day presentation as its foundational autonomous driving model. As for driver assistance features, Rivian stresses that they support the driver in operating the vehicle while the driver remains responsible for paying attention and maintaining control. The company says these features "do not replace their judgment or the need to remain attentive and in control of the vehicle at all times." Continuous supervision remains necessary when using them, unlike the driverless operation targeted by the robotaxi partnership.

In simulation tools used for training and testing, Third Dimension says its "SuperSim" platform builds realistic scenes directly from customer data. The company presents the approach as a way to reduce the time and cost of having graphics artists and engineers create scenes manually. It sums up its offering with the phrase "Build realistic scenes directly from your data," instead of relying on a large team to create them. According to the company's explanation of the platform, software can be replayed and its performance validated using different behaviors and objects within the scenes. Uses include testing rare or dangerous scenarios in a controlled environment. Third Dimension also explains that world models take actions as inputs and then simulate the environment's response. During training, this allows the outcomes of different decisions to be explored, alongside reconstructing the scenes in which the software is tested.

According to Rivian's roadmap, a limited rollout of "point-to-point" driving will begin later in the year the roadmap was published. The company then aims to make it widely available to customers in 2027 as part of driving assistance features that require continuous supervision. Commercial robotaxi operations under Uber's plan are targeted for 2028, according to the Associated Press. The planned launch includes San Francisco and Miami. The subsequent rollout plan targets 25 cities by 2031. The announced geographic scope of that expansion covers the United States, Canada and Europe.