Hyundai-backed AV company Motional gets back on the road

Motional is trying to have a Goldilocks moment in its quest to scale autonomous driving tech in a safe and affordable way.

Some approaches to autonomy, Motional execs contend, are effective but too expensive, due to costly and complex architectures that combine many different specialized machine learning models. Others, they say, are cost-effective end-to-end AI solutions but with performance limitations. Motional aims to get this balance just right, CEO and President Laura Major said during a presentation at Motional’s technical center in Las Vegas during CES 2026.

“We’re somewhere in the middle of these two approaches,” Major said. “It’s using the best from specialized models with a progression to mature these and grow the model size so it can generalize.”

This is the strategy behind Motional’s return to the global AV race after a hiatus. Leaders aim to launch a commercial driverless robotaxi service—enabled by a revamped AI-first autonomous driving system—in partnership with a major ride-hailing platform by year’s end. The company is starting out in Las Vegas, with plans to expand to Pittsburgh from there—and potentially into global markets and applications beyond ride-hailing in the future.

“We have a lot of plans to expand in many different directions,” Major said, “but first focused on robotaxi and achieving a safe commercial driverless service.”

History lesson: In 2020, auto tech supplier Aptiv and Hyundai Motor Group formed Motional (Aptiv has since significantly reduced its stake). The company put driverless vehicles on public roads in 2021, and in 2022 inked a 10-year deal with Uber (it also has an agreement with Lyft). In 2023, Hyundai started producing the electric Ioniq 5 robotaxi for Motional.

The Ioniq 5 has custom features geared toward driverless operations, such as automated doors, a passenger display screen in the back, cabin monitoring, and audio capabilities that enable communication with people inside and outside of the vehicle.

The company completed more than 130,000 customer rides on ride-hailing networks and has logged 2 million autonomous miles.

“In 2024, we saw that there was tremendous potential with all the advancements that were happening within AI. And we also saw that while we had a safe driverless system, there was a gap to getting to an affordable solution that could generalize and scale globally,” Major said. “So we made the very hard decision to pause our commercial activities, to slow down in the near term so that we could speed up.”

What’s next: Now Motional is back on the road. The company is launching a pilot program for employees that spans parts of Las Vegas, including the Strip (and, importantly, the designated pickup and drop-off zones outside hotels and casinos there), downtown Vegas, and the Town Square shopping district. For now, human safety operators remain in the vehicles, but Motional leaders aim to go fully driverless later this year.

Journalists got to experience Motional’s current capabilities during demo rides at CES that spanned an approximately 10-mile route. At no point during my ride did the safety operator intervene.

All in all, the trip was uneventful, save for one sudden stop after a vehicle traveling the opposite direction attempted to turn at the same time as the robotaxi. The vehicle was back on course within seconds and otherwise deftly handled dense highway traffic, congestion on the Strip, and pedestrian-heavy areas.

Motional’s experiences in Las Vegas and Pittsburgh, Major said, will help the company create a generalizable, scalable autonomous driving system.

“The reason we chose these two cities is because they represent two different city archetypes that we think are pretty representative of the various cities globally,” she explained. “Las Vegas is a modern city, has a grid-like structure, has a lot of multi-lane roads, a lot of dense traffic, a lot of lane-changing that has to happen.”

“Whereas Pittsburgh might look more like what you would see in Europe or in some other, older cities in the US,” she added. “It’s a historic city: windy roads, narrow streets, bridges, tunnels, a lot of interaction between pedestrians and vehicles.”

From there, Motional execs plan to leverage the company’s new AI-first solution to expand to new markets. Motional is integrating ML-based driving models and shifting to Large Driving Models.

Russell Ong, who previously worked for self-driving company Zoox and now leads product management at iMerit, which provides curated data for AI and ML applications, told us that one of the key differences between ML- and AI-based systems is that it can be difficult to understand why AI made the decisions that it did—and therefore to publicly validate its processes. “I would expect companies wanting to put robots on public roads to cover that base,” he said.

“We’ve been able to transition from a more classic robotic solution, all the way to a foundation model-based approach,” Major said. “This is really critical for two things. One is for generalizing more easily to new cities, new environments, new scenarios. And the other is to do this in a cost-optimized way. So for example, the traffic lights might be different in the next city you go to, but you don’t have to redevelop or reanalyze those. You just collect some data, train the model, and it’s capable of operating safely in that new city.”