Toronto-based autonomous trucking startup Waabi announced Tuesday that it has raised US$200 million in a Series B funding round led by Uber and Silicon Valley investment firm Khosla Ventures, as it aims to put driverless trucks on the road next year and expand its generative AI-powered testing model to new industries.

While Uber and Khosla are returning backers, the round includes a slew of new investors across varied industries, such as: chip giant Nvidia, Volkswagen’s parent company Porsche SE, investment firm HarbourVest and Ingka Investments, the investment arm of Ikea’s parent firm. Other participants include existing investors BDC Capital, Volvo Group Venture Capital and Toronto-headquartered Radical Ventures.

The wide swath of backers reflects Waabi’s search for the “best in class in terms of deep technology, AI automotive as well as in our logistics ecosystem,” CEO Raquel Urtasun told The Logic in an interview. The round is a vote of confidence that there is still room for growth in the autonomous-vehicle industry, despite high–profile accidents at Waymo and Cruise, and downsizing at startups like Open AI-backed Ghost Autonomy and Volkswagen-backed Argo AI.

Since Urtasun launched Waabi in 2021 with a US$83.5-million Series A round, interest in generative AI has soared. Waabi uses the technology in its training simulations, and Urtasun said the company’s AI model has enough general skills to be trained for other types of vehicles as well as humanoids and warehouse robotics.

That attracted Nvidia’s interest—as well as unlikely companies like Ikea, which has adopted other transportation and logistics technology, like Quebec-made electric vehicles from Lion Electric. Peter van der Poel, managing director of Ingka Investments, said in a statement that the investment reflects the Ikea parent’s goal to improve “safety, efficiency, and environmental impact” in the movement of its goods.

Waabi plans to expand its team in both Canada and the U.S. with the funding. 

Companies like Waabi signal how the industry is moving away from a “1.0” model of autonomous driving, said Urtasun, a University of Toronto professor who was previously R&D chief at Uber’s ATG autonomy division and an assistant professor at the Toyota Technological Institute at Chicago. Instead of being solely trained on data from droves of humans—who may or may not be good drivers—AI models also need to be able to reason through new situations, Urtasun said. 

That’s the approach the startup will seek as it prepares to remove drivers from the cabs of its commercial vehicles as soon as next year, with the hope that simulations will help diagnose issues and prove the vehicles’ safety in more situations before even hitting the physical road.

“When you are on the road, there is always that immediate reaction. But if you have a bit more time, you actually think about the consequences to make the right decision, then you end up with a much better and safer driver,” Urtasun said. 

“With this next generation, we are able to really incorporate that ability to reason which is so fundamental. And this enabled us to be much more data efficient.”