Waabi, the Toronto-based AI company building software to enable autonomous driving, has raised $1 billion in new funding and struck a major partnership with Uber to deploy at least 25,000 robotaxis on the ride-hailing giant’s platform.
The deal marks a significant expansion for Waabi, which until now has focused on autonomous trucking.
The funding consists of a $750 million Series C round led by Khosla Ventures and G2 Venture Partners, plus an additional $250 million milestone-based investment from Uber tied to the robotaxi deployment. The company says it is the largest fundraise in Canadian history.
Other investors in the Series C include Uber, NVentures (Nvidia’s venture capital arm), Volvo Group Venture Capital, Porsche Automobil Holding, BlackRock, Radical Ventures, and a subsidiary of the Abu Dhabi Investment Authority.
Waabi declined to disclose its valuation following the funding round. Toronto newspaper The Globe and Mail reported in December that the company was seeking a $3 billion valuation in the Series C round.
Waabi also declined to say where its Uber robotaxis would first be deployed or on exactly what timeline they would be rolled out.
Waabi represents a new breed of autonomous vehicle company—part of what some in the industry call “AV 2.0.” These companies use end-to-end AI models that learn to drive from vast amounts of data. Often a single AI model handles perception (understanding where the vehicle is on the road and what is happening around it); navigation (deciding what route to take); and action (deciding how to turn the steering wheel and whether to accelerate or brake).
This contrasts with earlier self-driving technology, such as that originally deployed by Alphabet company Waymo, which relied on extensive hand-coded rules, many different software programs and machine learning models, with each handling a single aspect of driving, as well as high-definition maps.
Uber has recently announced a slew of robotaxi deals with vehicle manufacturers and AV 2.0 startups. In many of those deals, Uber is providing the startups with funding, as it’s doing with Waabi. Earlier this month, Uber announced a tie-up with Nuro, another startup building software for self-driving, and Lucid Motors, which aims to put 20,000 Uber robotaxis on the roads, with the first robotaxi deployed this year.
Alongside that announcement, Uber invested $300 million into Nuro and Lucid. The ride-hailing company also has partnerships with self-driving startup Avride for robotaxis in Dallas and several other U.S. cities. And it has partnered with Waymo to allow passengers to hail Waymo self-driving cars through the Uber app in Austin and Atlanta. In 2024, Uber invested in U.K. AV 2.0 company Wayve as part of a partnership that also aimed to test Wayve’s technology in Ubers in London. Uber has a partnership as well with Chinese internet giant Baidu to test robotaxis in London and several other international markets.
Raquel Urtasun, the computer scientist who founded Waabi in 2021 and serves as its CEO, previously led Uber’s autonomous vehicle research lab. Uber has been involved with Waabi since its Series A venture funding round and already holds a seat on the startup’s board.
Previously, Waabi had been working on software that could operate autonomous trucks. In October, it announced the integration of its AI software into Volvo’s fleet of autonomous trucks, which provide autonomous freight delivery services on highways in Texas and at some mining and quarrying sites in Norway and Sweden. Volvo Autonomous Solutions also has a partnership with Uber’s Uber Freight service.
Currently, Volvo’s trucks that use Waabi’s software are using safety drivers in Texas. Urtasun said Waabi decided not to launch fully driverless trucking operations until the Volvo platform is fully validated—a decision she framed as prioritizing safety over speed. Volvo has said publicly that full validation is “just quarters away.”
Urtasun told Fortune that the expansion to robotaxis is in no way a pivot for Waabi. The company’s “physical AI platform” can generalize across different vehicle types, geographies, and driving conditions, and the exact same AI models that drive Waabi’s trucks will also power its robotaxis, she said.
“The model will be aware which vehicle it’s driving, but it will be the same model,” Urtasun said. “Think of us as humans—we are not switching our brain, but we know each vehicle we are driving.”
This approach stands in contrast to companies that have developed separate systems for different vehicle types. It also means that improvements made for trucking benefit the robotaxi system, and vice versa.
Although Waabi and Uber did not disclose a timeline for the Waabi-powered robotaxi rollout, Urtasun said it would happen “super fast. Much faster than anybody can think,” she said. “Much faster than you had traditionally seen on the robotaxi side.”
The robotaxi market is becoming intensely competitive. Waymo, owned by Google parent Alphabet, has been aggressively expanding beyond its original base in the San Francisco Bay Area. The company now operates in Phoenix, Los Angeles, Austin, and Atlanta, and has announced plans to launch in more than a dozen additional U.S. cities in 2026, including Miami, Dallas, Houston, Detroit, and Washington, D.C. It’s also planning its first international launches in London and Tokyo.
Tesla, meanwhile, launched a limited robotaxi service in Austin last June using its Full Self-Driving software. The service initially operated with human safety monitors in the passenger seat but began offering some fully driverless rides in January. Tesla’s approach, like Waabi’s, relies on end-to-end AI trained on camera data—though Tesla uses a vision-only system without the lidar sensors most competitors employ.
Wayve, the British company that has raised more than $1.3 billion from investors including SoftBank, Microsoft, and Nvidia, is also pursuing end-to-end AI. But unlike Waabi, Wayve has focused primarily on passenger vehicles and advanced driver-assistance systems rather than trucking.
Waymo itself has been experimenting with end-to-end AI models and is rebuilding its own self-driving technology stack around them, as Fortune reported last year. But the company continues to rely on a combination of lidar, radar, and cameras for commercial operations.
Waabi’s new funding, meanwhile, will go toward accelerating its commercial progress in trucking while also supporting the expansion into robotaxis, Urtasun said.
Vinod Khosla, founder of Khosla Ventures, said in a statement that Waabi’s technology is “a fundamental leap forward” in how driverless technology is being developed. “Their remarkable progress in autonomous trucking and rapid expansion into robotaxis demonstrates how their technology unlocks for the first time true scale in the real world,” he said.
