Lucid Group, Nuro, and Uber Technologies earlier this month announced a premium global robotaxi program created for the Uber ride-hailing platform. Expected to launch later next year in a major U.S. city, the experience is being developed for comfort, safety, and scale with a service combining the software-defined vehicle architecture of the Lucid Gravity, the Nuro Driver Level 4 autonomy system, and Uber’s global network and fleet management.

“Autonomous vehicles have enormous potential to transform our cities for the better,” said Dara Khosrowshahi, CEO of Uber. “We’re thrilled to partner with Nuro and Lucid on this new robotaxi program, purpose-built just for the Uber platform, to safely bring the magic of autonomous driving to more people across the world.”

With operations in 70 countries and an average of 34 million trips per day, Uber believes its platform has the scale to make the benefits of self-driving vehicles accessible and beneficial to people in many places. The company aims to deploy 20,000 or more Lucid vehicles equipped with the Nuro Driver over six years. The vehicles will be owned and operated by Uber or its third-party fleet partners and made available to riders exclusively via its platform.

As part of a deepening relationship with each partner, Uber plans to make multi-hundred-million-dollar investments in both Nuro and Lucid.

“This investment from Uber further validates Lucid’s fully redundant zonal architecture and highly capable platform as ideal for autonomous vehicles, and our industry-leading range and spacious, well-appointed interiors, as ideal for ridesharing,” said Marc Winterhoff, Interim CEO at Lucid. “This is the start of our path to extend our innovation and technology leadership into this multi-trillion-dollar market.”

The robotaxi will leverage Gravity’s advanced technology platform, redundant electrical and controls architectures, and long range, which are said to make it a good fit for use in a scalable robotaxi offering. The longer 450-mi (724-km) EPA estimated range of the Gravity means less frequent downtime for charging, minimizing costs, and maximizing vehicle availability.

The Nuro Driver, which will enable the Gravity to operate at Level 4 autonomy, combines automotive-grade hardware and AI-powered self-driving software designed for reliability and cost-efficiency at scale. It integrates an end-to-end AI model with safeguards for precision and reliability, which enables rapid adaptation to new environments, functions, and vehicle platforms—reducing deployment timelines. The hardware will be integrated into the Gravity on Lucid’s assembly line and vehicles will receive Nuro’s software when commissioned by Uber.

“Nuro has spent nearly a decade building an AI-first autonomy system that’s safe, scalable, and vehicle-agnostic, proven through five years of driverless deployments across multiple U.S. cities and states. said Jiajun Zhu, Co-Founder and CEO at Mountain View, CA-based AV startup Nuro. “By combining our self-driving technology with Lucid’s advanced vehicle architecture and Uber’s global platform, we’re proud to enable a robotaxi service designed to reach millions of people around the world.”

Nuro will lead the development and validation of the service’s safety case across dozens of categories using simulations, closed-course testing, and supervised on-road testing to verify that the robotaxi will operate safely. The first Lucid-Nuro robotaxi prototype is already operating autonomously on a closed circuit at Nuro’s Las Vegas Motor Speedway proving grounds.

 

Validating the shift

The deal with Uber and Lucid validates the expansion of Nuro’s business model announced in September 2024 to include licensing its AI-driven autonomy platform to automotive OEMs and mobility providers. The Nuro Driver will enable up to SAE Level 4 passenger vehicles and mobility platforms, adding to its goods-delivery capabilities.

The strategic move leverages the company’s years of delivery deployments. With more than one million autonomous miles completed across its fleet of R&D vehicles and zero autonomous at-fault incidents, the company says that its technology has demonstrated reliability and commitment to safety in real-world conditions.

The change in strategy has also been validated with more external investment. In April, Nuro announced it had raised $106 million so far in its ongoing Series E funding round, which values the company at $6 billion. The round included participation from new and returning investors as well as strategic partners. Investors include funds and accounts advised by T. Rowe Price Associates, Fidelity Management & Research Company, Tiger Global Management, Greylock Partners, and XN LP.

The company says the raise underscores investor confidence in its autonomy platform and its licensing-driven business model. With more than eight years of development and four years of real-world deployments in California and Texas, Nuro is one of the few companies to have successfully deployed driverless technology at city scale with no safety driver. By licensing its vehicle-agnostic, cost-effective technology, Nuro enables integration across commercial fleets, robotaxis, and personal vehicles—opening up broad applications for its AI-first self-driving system.

“We’re excited to see strong investor enthusiasm for our Series E,” said Zhu. “Our technology, years of experience with driver-out Level 4 deployments, and focus on licensing uniquely position us to help automakers, mobility platforms, and commercial fleets accelerate their autonomy roadmaps.”

The round brought Nuro’s total funding to $2.2 billion to date. The Series E funding will support Nuro’s next phase of growth—scaling its AI-based technology platform and advancing commercial partnerships.

 

Targeting all roads

Earlier this year, during CES 2025 in Las Vegas, Futurride was given a behind-the-scenes tour of Nuro’s LVMS proving grounds with the key execs driving the company’s tech ambitions. Tilo Schwarz, Vice President of Engineering for the Nuro Driver, leads the teams building the autonomy elements—such as mapping, localization, behavior planning, prediction, perception—plus the system design verification/validation and the software project management, making it all work together.

He joined Nuro in 2023 after 20 years at Mercedes-Benz, working on driver assistance systems, active safety, and Level 4 autonomous driving. The move came at a pivotal time as some of the big OEMs have scaled back their Level 4 efforts, he said. “But we are ready when they ask us to get it onto their vehicles,” previewing the Lucid collaboration.

Among the autonomous vehicles at the facility were the third-generation R3 delivery bots that the company is known for. However, the company’s shift of focus from commercializing the R3 to deploying its technology on all road vehicles has been in the works for a while. That strategy change was announced in September 2024, but “we’ve been working on it for more than a year prior,” according to Jenna Bott, Director of Brand & Marketing for Nuro.

“Now we try to serve all roads, all rides,” added Schwarz. “The key change we did with this shift is to realize that the software we built can drive any vehicle. You can drive a robot, you can drive a car, because fundamentally it’s the same thing.”

Nuro has already launched two driverless vehicles, the R2 and R3 delivery bots, “with no chase cars, which a lot of other companies haven’t done,” said Nicholas O’Neill, Director of Strategic Partnerships and Vehicle Technical Program Manager for Nuro. “It’s that integration portion that sets us apart and why I’m excited about where we’re going with the Nuro Driver—the ability to take our system, understand how to validate that it’s safe enough that you don’t need to have a safety driver or a car following. And that it’s adaptable to multiple applications as long as it’s a road that we’re driving on. That’s our target.”

The company’s pivot to offering its tech on any road vehicle was reinforced when Nuro engineers discovered that the technology they built worked well on its Toyota Prius-based development vehicles, “so why would you limit yourself to one vehicle type?” asked O’Neill. They “have since adapted it seven times already to different vehicle platforms. Most other companies have done it once,” he added.

The company’s expanded ODD (operational design domain) now includes highways down to Level 2 systems, but it is not unlimited yet.

“We cannot drive on snow, for example,” said Schwartz.

To further expand its Driver capabilities, Nuro has been conducting a nationwide data collection initiative, sending its Prius development vehicles to now more than 150 cities around the U.S. to gather data from various regions and weather conditions, including in inclement weather like snow, which has also been helpful for training.

 

Safety through redundancy

Safety is the most important measure of system success, according to O’Neill.

“You want to make sure before you put your vehicle on the road and run it autonomously that you’ve done all of the data collection necessary to prove to yourself that you’re safe,” he said.

Nuro’s AI-first approach to autonomy allows for rapid scaling and safe operation on public roads. The Nuro Driver leverages advanced AI across its autonomy stack, allowing it to confidently navigate urban and highway environments. Safety remains paramount, with redundant systems and a parallel autonomy stack helping to ensure reliable performance.

The latest Driver hardware for multiple vehicle types features a longer detection range to enable safe highway driving; “massively reduced” cost, unlocking new potential applications; and much improved power and thermal efficiency. The simpler architecture has fewer modules that are more easily adaptable to different vehicle types.

Enabling the Nuro Driver advances are automotive-grade hardware with AI-powered self-driving software using including solid-state lidar and Nvidia Drive Thor. Anchoring its “layering of safety” approach is an Arm Neoverse-based ASIL-D-rated chip design, according to O’Neill.

Like most AV companies, Nuro uses a range of sensor modalities and supporting systems for better performance and redundancy. Safety features based on eight years of Nuro Level 4 experience are microphones for siren detection, clearing systems for inclement weather, cleaning systems for removing dirt from sensors, and redundancy in key safety-critical systems.

“That’s the way I think most responsible companies approach autonomous driving,” O’Neill continued. “You don’t want to rely on just a couple [of sensors]. You want to have a broad capability to drive through the situations that may occur on any road, which is our goal is to be able to drive on all roads in the United States with any type of vehicle.”

The end goal is to build the most scalable AI driver, according to Bott. That means “trying to keep our bottom-line cost down, using automotive grade hardware, and being capital efficient as a company” so that Nuro’s Driver can be deployed on as many road applications as possible.