Aug 18, 2026 earnings call — Robotaxi and joint deployment expansion
Total revenues reached USD 36.2 million, a 69% increase year-over-year.
Revenue $0.0B (est $0.0B) · EPS $-0.14 (est $-0.15)
Key quotes
"Currently, we have secured over 4,000 vehicle commitments with Uber and other overseas partners."
"Our robotaxi fleet expanded to 2,000 vehicles, putting us on track to deliver 3,500 vehicles by year-end."
"PonyWorld 2.0 caught this difference automatically, and we quickly trained a new version of Virtual Driver that fits local habit perfectly, with very few engineers involved."
Analyst Q&A 7
Why Uber chose Pony in Europe
Q (Ming-Hsun Lee · BofA): Why did Uber pick Pony among other autonomous driving partners for its European rollout?
A: Management said Uber picked partners with reliable technology at scale and attractive cost economics. Pony showed real-world evidence: large-scale Gen-7 Robotaxi operations in China and Europe and competitive total cost per mile. They noted a 2000-vehicle contract and said fleet size could expand if performance and economics validate.
"Uber always looks for autonomous driving partners whose technology is reliable at scale, and also whose cost structure brings the attractive economics." — Jun Peng (CEO·CHAIR)
Joint deployment and asset-light model
Q (Tim Hsiao · Morgan Stanley): What is the strategy for the joint deployment model? How does the commercialization work under the asset-light model?
A: They said the joint deployment model speeds fleet expansion with high capital efficiency and creates a win-win across the value chain. The asset-light model involves three parties: Pony supplies Gen-7 Robotaxis and the Virtual Driver, a mobility platform brings demand, and an operating company handles fleet management and maintenance. Financially, Pony may earn sharing-based revenue or technology licensing fees, and the 4,000 vehicle commitments are a multiyear growth catalyst for 2026 and beyond.
"This is an asset-light model for Pony to expand our fleet." — Haojun Wang (CFO)
PonyWorld 2.0 and R&D efficiency
Q (Paul Gong · UBS): What makes PonyWorld 2.0's self-evolution different? How does it improve R&D efficiency, and would open-sourcing the world model hurt Pony's moat?
A: They said world model precision matters because traffic probabilities differ by city. PonyWorld 2.0 uses AI to drive the workflow instead of a human-led process, so engineers mainly verify and validate. This reduces engineering resources needed to enter new cities and they believe the ability to scale is a large moat not affected by open-source generative world models.
"In other words, without adding R&D resources, we can either enter many new markets at the same time, quickly achieving safe and smooth L4 autonomous driving." — Tiancheng Lou (CTO)
Domestic market outlook (China)
Q (Ming Chung · Citi): How should we think about Pony's domestic market outlook heading into the second half of the year?
A: They said domestic fleet expansion remains a significant part of rollout and they will focus on highest-value markets first, especially Tier 1 cities. Scale in hubs like Guangzhou and Shenzhen shortens wait time, boosts retention and daily revenue per vehicle, and helps amortize operational costs to improve UE margins. They plan to enter cities such as Changsha, Hangzhou, and additional Greater Bay Area cities this year.
"China itself represents a massive mobility market with over 10 million taxis and ride-hailing vehicles." — Jun Peng (CEO·CHAIR)
Operational efficiency in robotaxi operations
Q (Xiaoyi Lei · Jefferies): How is operational efficiency achieved (remote assistance, utilization, charging, maintenance)? How do these gains help expand and enter new cities?
A: They said efficiency comes down to fleet-to-staff ratio and reducing human involvement through technology. Robotaxis can autonomously navigate, locate chargers, and self-park at depots, so Pony needs three people for every 100 robotaxis versus one-to-one for traditional taxis. These gains lower operating costs per vehicle, let them expand without inflating overhead, and are packaged as standardized procedures and automation tools for partners.
"For example, when all of our robotaxis return to a depot, they require zero human assistance." — Tiancheng Lou (CTO)
L4 light truck progress
Q (Kai Xiao · CICC): Can you update us on new business initiatives, especially progress with the L4 light truck?
A: They said the L4 light truck fits their vision and shares strong synergy with Robotaxi technology and infrastructure. The light truck nearly shares 100% of Robotaxi technology, is jointly developed with CATL, and they have partnerships and orders with SF Express and China Post Technology with deployment schedules in place. Management expects more OEM and fleet-operator collaborations to drive scaling this year.
"The light truck almost shares 100% of our Robotaxi's technology and operational infrastructure." — Jun Peng (CEO·CHAIR)
Demo vs scaling and safety
Q (Yujing Ni · Everbright): What are Pony's views on the comment "The Demo is Only 1% Of The Work" about demos versus real scaling?
A: They said building a demo and scaling are different games because autonomous driving is a probability problem and safety must be proven with time and mileage. You cannot shortcut scaling by deploying thousands of cars overnight; regulators require step-by-step validation and leaders must iterate very fast. For Pony, they stated they have already checked both boxes and are focused on expanding into more cities and deploying larger fleets.
"Proving safety takes time and mileage, and you cannot just shortcut by dumping thousands of cars on the street overnight." — Tiancheng Lou (CTO)