Nearly two years ago, Motional was at the crossroads of autonomous vehicles.
The company, born out of a $4 billion joint venture between Hyundai Motor Group and Aptiv, has missed the deadline to launch a driverless robotaxi service with partner Lyft. It has lost Aptiv as one of its financial backers, prompting Hyundai to step up with the others investment of $ 1 billion to continue. Some layoffs, including a 40% restructuring deduction in May 2024, has whittled the company from its peak of around 1,400 employees to less than 600. Meanwhile, advancements in AI are changing how engineers develop technology.
Motional must develop or die. That rest all and choose option no.
Motional told TechCrunch it has rebooted its robotaxi plans with an AI-first approach to self-driving systems and promises to launch a commercial driverless service in Las Vegas by the end of 2026. The company has opened its robotaxi service – with a human safety operator behind the wheel – to its employees. It plans to offer that service to the public with an unnamed ride-hailing partner later this year. (Motional already has relationships with Lyft and Uber.) By the end of the year, human safety operators will be withdrawn from robotaxis and commercial driverless services will actually start, the company said.
“We see that there is tremendous potential with all the advances that are happening in AI; and we also see that if we have a safe and driverless system, there is a gap to get an affordable solution that can be common and global scale,” said Motional president and CEO Laura Major during a presentation at the company’s Las Vegas facility. “And we made the very difficult decision to pause our commercial activities, to slow down in the near term in order to accelerate.”
This means moving from a classical robotics approach to an AI-based model. Motional is never without AI. Motional’s self-driving system uses individual machine learning models to handle perception, tracking, and semantic reasoning. But it also uses other rule-based programs for other operations in the software stack. And individual ML models create complex web software, Major said.
Meanwhile, AI models originally built for language are beginning to be applied to robots and other physical AI systems, including the development of autonomous driving. The transformer architecture allowed for building large and complex AI models, which eventually led to the emergence, and skyrocketing use, of ChatGPT.
Techcrunch event
San Francisco
|
13-15 October 2026
Motional is looking for a way to combine these small models and integrate them into a single backbone, allowing for an end-to-end architecture. It has also maintained a small model for developers, which Major explained gives Motional the best of both worlds.
“This is critical for two things; One is to generalize more easily to a new city, a new environment, a new scenario,” he said. “And the other thing is to do this in a cost-optimized way. So for example, the traffic lights may be different in the next city, but you don’t have to rebuild or reanalyze. You just collect some data, train the model, and it can be used safely in that new city.”
TechCrunch got a first-hand look at Motional’s new approach during a 30-minute autonomous drive around Las Vegas. A single demo cannot provide an accurate assessment of a self-driving system. However, it can identify weaknesses and differences from previous iterations, and measure progress.
Progress what I saw as the Hyundai Ioniq 5 I rode in autonomously navigated from Las Vegas Boulevard and went to the pickup and drop-off area of the Aria Hotel. This busy area is famous in La Vegas and my experience was no different as autonomous vehicles slowly circle taxis that stop and unload passengers, change lanes, then come back again, passing dozens of people, giant flower pots, and cars on the street.
Motional previously operated a ride-hailing service in Las Vegas with partner Lyft using vehicles that would autonomously handle part of the ride. The hotel’s parking and valet area and application pickup area were never part of the operation. A human safety operator, always behind the wheel, will take over to navigate the parking lots or busy pickups and point-off hotel lobby.
There is still progress to be made. The graphics displayed for the riders in the vehicle are still under development. And while there were no distractions during the demo — meaning human safety operators took over — the vehicle took time to ride in a double-parked Amazon delivery van.
Still, Major says Motional is on the right track to deploy safely and cost effectively. And Hyundai’s majority owner is in it for the long haul, he said.
“I think the real long-term vision, you know, for all of this, is to put Level 4 in people’s personal cars,” said Major, referring to the term that means the system handles all driving without the expectation of human intervention. “Robotaxis, it’s the number one stop, and it has a big impact. But in the end, I think any OEM likes to integrate that into the car.