Kodiak’s CEO says making trucks drive themselves is only half the battle

This year is expected to be a big year for self-driving trucks. In addition to Aurora’s plan Deploy hundreds of large autonomous excavators and The reference expands on roboticsI got it too Kodiak AI It aims to launch a fully driverless long-distance freight operation by the end of 2026. While robotaxis may still be grabbing all the headlines, self-driving trucks are making inroads of their own, slowly but surely.

But in a recent interview, Don Burnett, CEO of Kodiak AI, said deploying self-driving trucks is only half the battle. While most of its competitors fuss over details like artificial intelligence, perception and cut-offs, Kodiak plans for the reality of running a business, he said. This includes answering important questions such as who owns the trucks, how much uptime is required, and what will ultimately be shipped.

“So you start to realize that it’s not just a truck operating safely on the road…it’s a risk on the table,” Burnett said. “What really matters to customers is how efficiently and effectively I get that truck in and out of my business…and everything in between. And no one talks about that either.”

Kodiak AI (formerly Kodiak Robotics) was founded in 2018 by Burnett, a veteran of Google’s self-driving car project (now called Waymo), and Buzz Eichel. The company is developing self-driving trucks for highway and industrial uses. As well as the defense industry. In 2025, the company’s trucks began delivering orders without a driver Atlas Energy Solutions is in the oil-rich Permian Basin of western Texas and eastern New Mexico20 driverless trucks are now operating there. Kodiak AI went public through a reverse SPAC merger in September 2025.

Now the company is finally ready for the open road. Burnett said Kodiak works across multiple sectors, with a focus on industrial and off-road trucking, which he sees as a big opportunity compared to traditional on-road autonomous driving. He describes these environments as “unstructured,” because they are more complex and unpredictable. In this way, he says his trucks are better prepared for more “regulated” environments, such as highways.

“We plan to recall the driver by the end of the year,” Burnett said. “Remember, a product is only valuable if it is driverless.”

But first, Kodiak needs to complete its safety case. This involves collecting large-scale data, driving virtually in a simulated world, and creating a detailed risk mitigation plan. Burnett said his team’s background at Waymo helped influence its tough approach to safety.

Kodiak also takes a different approach to its business model than some other companies. Unlike competitors who expected OEMs to offer self-driving-ready trucks, Burnett said Kodiak is developing an aftermarket solution in partnership with companies like Roche Industries and Bushallowing them to produce fully compatible car-class trucks and scale them more effectively once the technology is ready. As such, the 20 trucks Kodiak has delivered so far are owned and operated by its customers, not Kodiak itself.

This is where Kodiak differs from its peers, and even Waymo. Burnett says that when a customer owns a vehicle, they care about key metrics like usage, uptime, maintenance and revenue at all times. He says this creates a much higher level of reliability and operational performance.

“When a customer owns the car, they have to work,” Burnett said. “Customers are going to expect the truck to run all the time. So you have to get that far before you can really sell the truck to the customer.”

When an autonomous vehicle developer owns the truck, rather than the customer, they can manage their own deployments without worrying about real-world functionality.

“Those are their own trucks,” he says of his competitors. “So, if they’re only working one day a week, or they’re only working five hours in one day, no one cares…they’re still without a driver and they can still claim victory. But that’s never going to work with the customer. Like, that’s not a real product.”

Burnett can be blunt when talking about his competitors. When asked if Kodiak would expand its product portfolio to include taxis, as Wabi did recently, he replied: “Did Wabi have a product before it existed?” He believes Kodiak outperforms its competitors in terms of real-world deployment and operational accuracy. He suggests that many self-driving car companies emphasize technology and great visuals, but have not crossed the harder threshold of offering a usable and customer-owned product:

“They make great videos… and they have attractive images,” he says of his competitors.

He asserts that most companies haven’t solved what he calls the third pillar: making autonomy actually usable in real workflows. This includes integrating self-driving trucks into customers’ operations, handling the complexities of transportation and delivery operations, and providing monitoring and communication tools. The rest of the package focuses on driving performance, completely ignoring system integration. Kodiak AI does both.

“No one talks about this,” he says.