Waabi Uses Generative AI for Driverless Autonomous Trucking

Artificial intelligence is transforming the transportation industry, helping drive advances in autonomous vehicle (AV) technology.

Waabi, a Toronto-based startup, is embracing generative AI to deliver self-driving vehicles at scale — starting with the long-haul trucking sector.

At GTC in March, Waabi announced that it will use the NVIDIA DRIVE Thor centralized car computer to bring a safe, generative AI-powered autonomous trucking solution — the Waabi Driver —  to market.

As the company plans the launch of fully driverless operations next year, Waabi is reinvigorating the industry with a self-driving solution that’s capital-efficient, can safely handle new scenarios on the road and ultimately scales commercially.

Waabi is developing on NVIDIA DRIVE OS, the company’s operating system for safe, AI-defined autonomous vehicles.

The innovative startup has pioneered an approach that centers on the combination of two generative AI systems: a “teacher,” called Waabi World, an advanced simulator that trains and validates a “student,” called Waabi Driver, a single, end-to-end AI system that’s capable of human-like reasoning and is fully interpretable.

When paired together, these systems reduce the need for extensive on-road testing and enable a safer, more efficient solution that is highly performant and scalable.

“We are excited to have a deep collaboration with NVIDIA to bring generative AI to the edge, on our vehicles, at scale,” said Raquel Urtasun, founder and CEO of Waabi.

Generative AI accelerates the development of AVs by “providing an end-to-end system where, instead of requiring hundreds of engineers to develop a system by hand, it provides the ability to learn foundation models that can run unsupervised by observing and acting on the world,” Urtasun added.

Waabi’s collaboration with NVIDIA is one in a series of milestones, including the company’s $200 million Series B round with participation from NVIDIA, its work with logistics company Uber Freight, the launch of its first commercial autonomous trucking routes in the U.S., and the opening of a trucking terminal near Dallas to serve as the center of the company’s operations in the Lone Star state.

“What we’re building for autonomous vehicles — combining generative AI-powered simulation with a foundation AI model purpose-built for acting in the physical world — will enable faster, safer and more scalable deployment of this transformative technology around the world,” Urtasun noted on the company’s website.

Listen to Urtasun’s talk at GTC for more on the company’s work on using generative AI to develop autonomous vehicles.