From Gen-4 heavy-duty trucks to L4 urban delivery vehicles, shared technology, automotive-grade production and industry partnerships are shaping the next phase of Pony.ai's autonomous freight business
GUANGZHOU, CHINA, Aug. 6, 2026 /PRNewswire/ -- At a media briefing on August 3, He Xing, Vice President of Pony.ai and Head of the company's Robotruck business, discussed how nearly a decade of technology development in L4 autonomous driving is beginning to support broader commercial deployment in freight transportation.
Over the next two to three years, Pony.ai expects 500 to 1,000 Gen-4 autonomous heavy-duty trucks to be deployed across three primary scenarios in China: long-haul freight, bulk commodity transportation and port logistics. Pony.ai also expects light-duty trucks to scale faster, with a longer-term goal of reaching 100,000 L4 autonomous light-duty trucks by 2030.
The targets reflect several developments coming together: a more mature autonomous driving system, lower hardware costs, automotive-grade redundant vehicle platforms and deeper collaboration with vehicle manufacturers and logistics operators.
Why freight, and why now
Pony.ai began developing autonomous trucks in 2018. The first vehicles were largely hand-built prototypes. Subsequent generations moved progressively closer to automotive-grade production through partnerships with truck manufacturers. The question gradually shifted from whether the technology could work to where and under what conditions, it could create the most operational value.
Road freight presents a clear need. The industry faces persistent structural pressures, including a shortage of qualified heavy-duty truck drivers, an aging workforce and sharp fluctuations in demand during peak seasons. Long hours, overnight driving and demanding routes can also increase fatigue-related safety risks. L4 autonomy can help supplement freight capacity, particularly on repetitive routes and during overnight or peak-demand periods that are difficult to staff, while supporting safer, more consistent operations.
Turning that potential into scaled operations, however, requires more than technical capability. Autonomous driving systems must also be safe, reliable, ready for automotive-grade production and commercially viable at scale. For Pony.ai, the timing of large-scale deployment has therefore also depended on bringing down the cost of the autonomous driving system.
"We had been waiting for the right moment," He said. "Our truck technology had already reached a high level, but the cost of building an L4 Robotruck remained high. The reduction in ADK costs benefited both heavy- and light-duty trucks. That is why we did not rush into large-scale production earlier."
Pony.ai's Gen-4 autonomous heavy-duty truck has reduced autonomous driving hardware costs by approximately 70% compared with the previous generation. Developed for automotive-grade mass production, the vehicle is designed for a service life of 20,000 operating hours or up to 1 million kilometers.
The Robotruck business has already moved beyond technology testing. As of November 2025, Pony.ai operated a fleet of around 200 trucks and had transported more than 1 billion ton-kilometers of freight. In the first quarter of 2026, Robotruck services generated US$10.2 million in revenue, up 31% from a year earlier, driven primarily by the expansion of commercial operations.
Two vehicle platforms across the freight network
Pony.ai's current Robotruck strategy covers both heavy-duty trucks and light-duty trucks, reflecting the different roles they play across the freight network.
Heavy-duty trucks are designed primarily for transportation between logistics hubs, including long-haul highway freight, bulk commodity routes and port transportation. Pony.ai's Gen-4 models are based on battery-electric platforms and support both single-vehicle autonomous operation and L4 platooning, depending on the requirements of each operating environment.
Production of the Gen-4 heavy-duty trucks is now underway. Vehicles are expected to roll off the production line in batches and enter commercial service across several use cases over the coming months.
Shenzhen's Mawan Port will be among the first deployment sites. Pony.ai has secured a project there and expects to deploy dozens of Gen-4 Robotrucks for commercial operations. Ports represent one of the three priority scenarios for the platform, alongside highway freight and bulk commodity transportation in western China.
Light-duty trucks address a different part of the logistics chain. They are commonly used between urban distribution centers, retail stores, delivery outlets and cold-chain facilities—environments that overlap substantially with the complex urban road conditions in which Pony.ai's Robotaxis already operate.
Pony.ai introduced its first L4 autonomous light-duty truck in April 2026. Co-developed with CATL and built on CATL's Kunshi Chassis Platform, the vehicle uses automotive-grade components and a fully redundant safety architecture. It offers approximately 18 cubic meters of cargo space and is intended for both urban and intercity freight.
The first vehicles have now entered intensive road testing in operating environments provided by logistics partners. Initial use cases include express delivery, retail distribution and food and beverage cold-chain logistics. Pony.ai plans to pursue the regulatory approvals required for fully driverless operation as testing and validation progress.
Based on current operating assumptions, Pony.ai estimates that fully driverless light-duty trucks could reduce per-kilometer operating costs by 40% to 50% compared with conventional human-driven operations. The vehicle can also carry 2.6 times the cargo volume of mainstream low-speed autonomous delivery vehicles, while operating at speeds suitable for regular urban and intercity roads.
One Virtual Driver across vehicle types
Pony.ai's approach is built around applying the same underlying Virtual Driver technology across Robotaxis, heavy-duty trucks and light-duty trucks.
The light-duty truck uses the same core technology stack as Pony.ai's Gen-7 Robotaxi. Because the two platforms operate in many of the same urban environments, they can also share supporting infrastructure and operating capabilities, including charging, ground support, service centers, fleet management and remote assistance. Pony.ai estimates that the overall technological and operational synergies between the two platforms exceed 90%.
Heavy-duty trucks require more vehicle-specific adaptation. Their size, weight, mechanical structure and longer braking distances create different control requirements, while highway and bulk commodity routes introduce operating conditions not commonly encountered by passenger vehicles. Even so, the core capabilities used to understand traffic, interact with other road users and make driving decisions draw on the same underlying technology and development methodology.
Data and operating experience from the different vehicle platforms also contribute to a shared development loop. PonyWorld 2.0, Pony.ai's proprietary world model, is designed to identify areas where the Virtual Driver requires further improvement, guide targeted data collection and support more efficient training and evaluation.
"Autonomous driving has to progress step by step—from technology driving product development, to the product enabling a business model, and ultimately to that model reshaping the industry," He said.
The shared safety architecture is equally important. Pony.ai's current Robotaxi, heavy-duty truck and light-duty truck platforms use redundant systems covering steering, braking, communication, power supply, computing and sensing. This fail-operational design allows a vehicle to maintain core driving functions and select an appropriate location to pull over safely if certain hardware or software components fail.
A partner-led route to scale
Scaling autonomous freight requires more than producing vehicles. It also requires access to freight demand, established operating networks, maintenance capabilities and infrastructure such as logistics hubs and charging facilities.
Pony.ai has therefore built its Robotruck business around partnerships with vehicle manufacturers and logistics operators. Its Gen-4 heavy-duty trucks were developed in collaboration with manufacturers including SANY Truck, while the light-duty truck was co-developed with CATL. Pony.ai also works with Sinotrans across long-haul freight and urban logistics scenarios.
The commercial model can vary depending on the maturity and requirements of a project. In some earlier-stage deployments, Pony.ai participates more directly in vehicle ownership and freight operations through a Transportation-as-a-Service, or TaaS, model. This allows the company and its partners to validate operating performance and unit economics in real commercial environments.
As the market matures, Pony.ai expects partner-led deployment under an Autonomous Driving-as-a-Service, or ADaaS, model to play a larger role. Under this model, vehicle manufacturers produce the trucks, logistics partners own and operate the fleets, and Pony.ai provides its Virtual Driver and related technical services. Some port projects are already beginning to adopt this approach.
"We are not here to run e-commerce or postal services ourselves," He said. "Our role is to become a partner to logistics companies and integrate into the systems they already use to serve their customers."
This structure allows each participant to focus on its established strengths: vehicle manufacturers on automotive-grade production and sales, logistics companies on freight demand and fleet operations, and Pony.ai on autonomous driving technology.
The next phase
The next phase of Pony.ai's Robotruck business will focus on ramping up production of the Gen-4 heavy-duty truck and deploying it in commercial projects, while completing the testing and regulatory work required to deploy the light-duty truck at scale.
In heavy-duty trucking, Pony.ai will initially focus on highway freight corridors, bulk commodity routes and ports. The company is also exploring an innovative model for highway transportation, which could simplify trailer handoffs between autonomous highway operations and human-driven first- and last-mile delivery.
For light-duty trucks, the immediate focus is to work with logistics partners to validate operations in express delivery, retail distribution and cold-chain transportation. The ability to operate overnight when drivers are more difficult to recruit and fatigue-related safety risks are higher could become an early commercial use case.
Pony.ai also sees potential for autonomous trucks in overseas markets, particularly at ports and other well-defined logistics sites where driver shortages and labor costs strengthen the economics of automation.
For Pony.ai, the objective is not simply to place more autonomous trucks on the road. It is to build a repeatable operating model in which technology, vehicles, infrastructure and freight demand can scale together.
