(Image credit: Hyundai/Motional)

Stroll a parking lot in Las Vegas and you are bound to bump into a company on the brink of making a major autonomous driving breakthrough. Yet not many go on to survive the difficult early years and emerge out of the other side with a successful business.

Fortunately, this year’s CES 2026 was a backdrop for Hyundai and Motional to celebrate, as they dropped the silk sheet on the fruits of their partnership and announced that they plan to run a driverless service in Las Vegas later this year.

It hasn’t been an easy path though, with Motional (formerly a joint venture with automotive tech company Aptiv) pausing its operations in 2024 to take stock after six years of perfecting its autonomous recipe.

“We’ve given 130,000 rides to the public on the Lyft and Uber network, including delivering food via Uber Eats, and we’ve driven over two million autonomous miles with zero at fault incidents,” explains Laura Major, Motional CEO, at the company’s Las Vegas nerve centre.

“But a lot has changed in the AI space in recent years and we wanted to focus on accelerating our path to advanced AI technology,” she says.

(Image credit: Hyundai/Motional)

Motional’s CEO says that during this downtime the company has been able to transition from more classic robotic solutions, which are time-consuming and expensive, to harnessing neural networks, large language models and vision-language-action models that help generalize new cities, environments and scenarios to cut costs and speed up deployment.

On top of this, Hyundai increased its involvement with the business in 2020, becoming the majority owner of Motional, pumping in additional funds and vowing to supply its purpose-built Ioniq 5 Robotaxi to the fleet for testing and deployment.

Sign up for breaking news, reviews, opinion, top tech deals, and more.

The stars aligned and now Motional feels that it is ready to take on the might of Waymo, Tesla and Zoox, which also has a handful of its autonomous pods running around the famous Las Vegas strip.

Purpose-built Johnny Cab

(Image credit: Hyundai/Motional)

Hyundai’s Ioniq 5 robotaxi is assembled at its cutting-edge smart factory in Singapore, where robotic Spot dogs from Boston dynamics oversee quality control. The key difference to many competitors is that the vehicle, complete with massive sensor and Lidar suite, is made on a production line at scale.

Waymo is about to introduce the Ioniq 5 to its fleet in order to replace its aging Jaguar I-Pace cars, but Motional will be first to hit the streets with Hyundai’s award-winning electric vehicle.

Like the regular passenger cars, the fully autonomous Ioniq 5 comes with a steering wheel and space inside for four passengers to sit comfortably. The only real differences are the automatically opening and closing doors, the screens for rear passengers to chart their autonomous journeys and a number of buttons dotted around the cabin and on the exterior bodywork that allow customers to dial into a call center if help is needed.

(Image credit: Hyundai/Motional)

My test drive started in a quiet business district near Harry Reid International Airport and, despite a safety operative in the driver’s seat, I was encouraged to commence the journey like a customer would. So I tapped on the screen to begin the journey and the doors silently closed.

The route towards the Las Vegas strip is relatively simple, but the Motional Ioniq 5 didn’t hang around. It merged in turn and kept up with the speed of traffic, even pulling a fairly aggressive move in order to get in the correct lane to turn right.

But the really impressive stuff took place on the Vegas strip, which is notoriously busy and jam-packed with unpredictable behaviors. The Ioniq 5 navigated this with ease and even pulled into a hotel valet parking lot just to show off its skills.

Previously, when Motional was operating early rides with Lyft, the safety driver would interrupt at this point and help steer the vehicle through the most complex parts – like a pilot manually lands a plane.

This time, the car was left to fend for itself, allowing cars to pull out and even stopping safely to let a bellhop and a pedestrian cross in front. Above all else, the braking was smooth and it didn’t make any worrying sudden movements.

(Image credit: Hyundai/Motional)

The journey was almost faultless but I did experience one disengagement – where the safety operator had to take over proceedings. On the way back to Motional’s HQ, a vehicle essentially undertook the car on the left and then cut in front of it at some traffic lights.

Motional’s robotaxi then decided it wanted to pass the now stationary car on the right, but that would have forced us into a lane that could only legally turn right. Our safety driver had to take control for a second to ensure it didn’t.

For instances like this, Motional has a bustling operation room at its Vegas HQ that allow a teleoperator to make decisions and remotely send that to a car.

But Adam Griffin, Vice President of Operations and Head of Safety at Motional, claims they are seeing fewer and fewer circumstances where the team has to get involved.

AI cranks up the pace

(Image credit: Hyundai/Motional)

It is exactly these sort of “challenging edge cases” that Laura Major claims AI has helped the company tackle in recent years.

Part of the work has been creating what the company calls an “Omni Tag” process that uses large language models and vision-language models to search the reams of video footage Motional’s fleet has captured.

“We can find those critical scenarios that we can train on and improve our performance in those specialized tasks,“ Major explains. The example she gives is a car coming across a rickshaw for the first time and not really knowing how to behave. Now, Motional engineers can easily pull up examples of rickshaw encounters and train the model accordingly.

“Omni tag allows us to go from what used to take large teams of data scientists to curate and find these valuable data sets. Now we can do that automatically. So it used to take us months to curate these data sets. Now we can do it in hours and minutes,” Major adds.

(Image credit: Hyundai/Motional)

It is this unique mix of a classical robotics background and an advanced AI approach that Major thinks puts Motional in a great place for success in a growing Robotaxi landscape.

But the company doesn’t want to stop at unleashing a fleet of autonomous taxis with a yet-to-be-named ride-hailing partner, as Major admits that the end goal is to perfect this technology so it can be made available for passenger vehicles and public consumption.

Seeing as Hyundai doesn’t even have autonomous driving tech that can match Tesla’s Full Self-Driving or Ford’s BlueCruise Level 2 systems yet, the huge investment in Motional could soon pay off, as it could end up jumping straight to Level 4.

Leon has been navigating a world where automotive and tech collide for almost 20 years, reporting on everything from in-car entertainment to robotised manufacturing plants. Currently, EVs are the focus of his attentions, but give it a few years and it will be electric vertical take-off and landing craft. Outside of work hours, he can be found tinkering with distinctly analogue motorcycles, because electric motors are no replacement for an old Honda inline four.

You must confirm your public display name before commenting

Please logout and then login again, you will then be prompted to enter your display name.