Strategy for Building Moats in Large-Scale Operation of Robotaxi Enterprises

The Robotaxi industry is at a core turning point from technological competition to operational competition. Relying solely on algorithmic and technological advantages is no longer sufficient to support long-term market competitive positions; large-scale and stable operation capability is becoming the key moat that determines the success or failure of enterprises.

As the fleet size expands, the fixed cost allocation effect is significantly enhanced. Data shows that when the Robotaxi fleet size reaches 1000 vehicles, operations can reach the break-even point; after crossing this threshold, the cost per additional vehicle will be lower, the gross profit margin will be higher, and it will enter a positive self-sustaining stage [4]. Pony.ai has achieved positive unit economic model in Guangzhou; after excluding fixed costs, scale expansion can achieve overall profitability [3].

The continuous reduction of the human-vehicle ratio for cloud-based remote driving is a key driver of cost reduction. With technological progress, on-board safety officers are gradually transformed into remote safety officers; the decrease in the safety officer-to-vehicle ratio can significantly reduce safe operation costs [1].

Robotaxi platforms have typical network effect characteristics. When the fleet size reaches 100,000 vehicles by 2030, Robotaxi will occupy 5%-10% of China’s mobility market, generating strong network effects and industry driving forces [4]. More vehicles cover more areas, attracting more users; more users generate more orders, further improving vehicle utilization efficiency.

Massive data generated from large-scale operations feeds back into technological iteration. Apollo Go’s global autonomous driving total mileage has exceeded 240 million kilometers, equivalent to circling the Earth 6,000 times, of which fully unmanned driving mileage reaches 140 million kilometers [2]. This data provides valuable learning samples for algorithm optimization.

Ruqi Mobility has built a three-level operation and maintenance network consisting of Robotaxi quick response sites, Robotaxi maintenance stations, and Robotaxi hub centers, covering functions such as charging, maintenance, data collection, and dispatching [5]. This systematic operation and maintenance capability is the infrastructure guarantee for large-scale operations.

Integration in the middle of the industrial chain promotes the close combination of vehicle manufacturers, autonomous driving technology companies, and mobility service operators; this cooperation model accelerates the large-scale deployment of the Robotaxi market [1].

Ruqi Mobility has become the first mobility platform to launch commercial hybrid operation of manned ride-hailing and Robotaxi services, effectively reducing operational risks and improving service continuity [5].

Apollo Go’s fully unmanned order proportion increased to 80% in October 2024, reflecting a progressive safety officer removal strategy [4].

Apollo Go’s cooperation with Uber adopts the “technology platform + local operation giant” model; over 1,000 Apollo Go unmanned vehicles will be connected to Uber’s huge network in Asia, the Middle East, Europe, and Oceania, significantly reducing policy risks and operational costs of market entry [2].

Based on the “model” effect of profitable fleets in Guangzhou, Pony.ai has proven the feasibility of its business model to vehicle asset holders and operators, attracting them to join the ecosystem in a light asset model, and has started cooperation with partners such as Sun Mobility and Shenzhen Xihu Group [3].

Hardware costs gradually decrease with industrial scale. Taking Baidu as an example, the unit cost of Apollo Go’s sixth-generation unmanned vehicle has dropped from one million yuan to 204,600 yuan [2]. The total cost of Pony.ai’s seventh-generation autonomous driving kit has decreased by 70% compared to the previous generation [5].

The unit shipment price of lidar has dropped from 14,317 yuan in 2022 to 3,917 yuan in 2024, and the cost of domain controllers is also continuously optimized [3].

2025 may become the turning point for Robotaxi to move towards profitability. Apollo Go plans to achieve break-even in Wuhan by the end of 2024 and fully enter the profit period in 2025; Pony.ai expects to achieve single-vehicle operational break-even (gross profit turning positive) in 2025 [4].

Pony.ai plans to grow its fleet size from 1,000 to at least 3,000 vehicles by the end of 2025, and reach the 100,000-vehicle level in the medium to long term by 2030 [3].

The Middle East market has become a key area for intelligent driving companies’ overseas plans. WeRide has cooperated with Uber in Abu Dhabi, and Pony.ai has received a US$100 million investment from Saudi Arabia’s NEOM and its investment fund [4].

Apollo Go has been deployed in 22 cities including Beijing, Shanghai, Wuhan, Shenzhen, Hong Kong, Dubai, and Abu Dhabi, demonstrating global deployment capabilities [2].

In 2024, 20 cities were selected as “vehicle-road-cloud integration” pilots, promoting the leap from testing to large-scale operations [1].

The policy system is increasingly improved, providing institutional guarantees for Robotaxi commercialization.

The competitive logic of the Robotaxi industry is shifting from “technological advancement competition” to “large-scale operation capability competition”. Technology is no longer the only differentiating factor; capabilities in operational dimensions such as operational efficiency, cost control, network effects, and ecological collaboration are becoming new moats.

Enterprises need to focus on building the following core capabilities while maintaining technological competitiveness:

1. Refined Operation Management
   : Optimize vehicle dispatching, route planning, and charging strategies through data analysis.
2. Standardized Service Process
   : Ensure service quality and consistent user experience under large-scale operations.
3. Ecological Cooperation Network
   : Establish deep cooperative relationships with vehicle manufacturers, component suppliers, local governments, and mobility platforms.
4. Rapid Replication and Expansion Capability
   : Establish standardized urban deployment and operation models to support rapid expansion.

Only enterprises that truly master large-scale and stable operation capabilities can stand out in the marathon competition of Robotaxi and achieve the transformation from technological leaders to industry leaders.

[1] Toubao Research Institute. Enterprise Competitive Landscape: 2024 Robotaxi Toubao Entry Report Series. November 2024.
[2] Baidu World 2025 Conference. Apollo Go’s Weekly Fully Unmanned Orders Exceed 250,000, Cumulative Services Exceed 17 Million Times. November 2025.
[3] Pony.ai. From Small-Scale Profitability to Global Expansion: Pony.ai Discloses Complete Robotaxi Business Logic. December 2025.
[4] Soochow Securities. Robotaxi Siege War: Land Grabbing but Not Profitable. November 2024.
[5] Soochow Securities. Robotaxi is Reshaping the Automotive Mobility Market. September 2025.
[6] Bank of America Securities. In-depth Report on Autonomous Driving: Unmanned Ride-Hailing Market Size Can Reach Trillions. 2024.