The evolution from L2 to L4 is a "false proposition" Pony AI's Robotaxi model commercialization and mass production are accelerating

Despite being listed on NASDAQ for nearly a month, Pony AI (PONY.US) remains a hot topic in the 2024 U.S. stock market's autonomous driving industry, with a total fundraising amount close to $452 million.

Pony AI has become a focal point in the industry, largely attributed to its provision of safe, advanced, and reliable full-stack autonomous driving technology, as well as its dual development paths of automation and commercialization. The prospectus shows that the company is committed to developing a "virtual driver" suitable for various vehicle models and scenarios, with its technology widely applied in three major areas: autonomous ride-hailing services (Robotaxi), autonomous trucks (Robotruck), and intelligent driving for passenger vehicles (POV).

In terms of L4 autonomous driving technology and scale, Pony AI is globally leading and has already achieved commercial operation of fully unmanned Robotaxi ahead of others.

Perhaps sensing the enormous potential of this market, Pony AI took the lead in exploring the commercialization of fully unmanned Robotaxi. According to data, it is expected that by around 2030, Robotaxi services will enter a mature commercialization stage, with the global market size expected to reach approximately $39 billion. According to Frost & Sullivan's forecast, the global Robotaxi market size is expected to be around $290 million in 2025, and this figure is expected to soar to $6.66 billion by 2030, with a compound annual growth rate of 195.6%. In the Chinese market, it is expected that by 2030, the Robotaxi market size will reach $352.6 billion, with a compound annual growth rate of 199.6%, accounting for more than half of the global market share.

Facing this vast market, Pony AI began its deep cultivation in the autonomous driving field from 0 to 1 in 2016, launching the first Robotaxi service in China in 2018—PonyPilot APP. Over time, it has obtained permits for unmanned driving services in the four major first-tier cities: Beijing, Shanghai, Guangzhou, and Shenzhen. In Beijing, Guangzhou, and Shenzhen, the company has initiated fully unmanned commercial charging operations for Robotaxi. Currently, Pony AI has deployed over 250 Robotaxi nationwide, with a total autonomous driving test mileage close to 40 million kilometers, of which nearly 4 million kilometers are unmanned test mileage. This data lays a solid foundation for the scaling and unmanned operation of autonomous driving services.

Regarding the commercialization of Robotaxi, at a media communication meeting on December 19 in Nansha, Guangzhou, Zhang Ning and Mo Luyi, Vice President and head of Robotaxi business at Pony AI, told Zhitong Finance that Pony AI has met the conditions for large-scale commercialization domestically, which mainly include: first, the technological conditions are mature, with the company possessing globally leading unmanned technology; second, policy support, as other first-tier and non-first-tier cities have successively introduced policies for Robotaxi on the road; third, the capability for large-scale production, with collaborations with Toyota, BAIC, and GAC for mass production, and the seventh generation of Robotaxi is expected to roll out the first batch of thousands of units in the next two years; Fourth, the ability to scale operations has accumulated a certain number of cognitive users, and the fleet size and autonomous driving mileage, which are among the top in the industry, make users feel more at ease; fifth, there is deep cooperation in the industrial chain, such as collaboration with sensor and chip companies, as well as mobility and logistics platforms, to jointly promote industry progress and development.

Thanks to a multidimensional layout, the commercialization of Robotaxi has also accelerated the performance release of Pony AI. In the first half of 2024, Pony AI's fully autonomous Robotaxi had an average daily order volume of over 15 per vehicle, and the revenue from the autonomous driving mobility service segment grew by 86% year-on-year, marking a key milestone in the large-scale commercialization of Robotaxi. During this period, the company's overall revenue increased significantly by 101.2% year-on-year. As the scale grows, the company's operating expenses have also continued to decline, with sales, general, and administrative expenses decreasing by 23.9% and 8.8% year-on-year in 2023 and the first half of 2024, respectively, indicating that Pony AI has achieved cost reduction by improving operational and marketing efficiency.

At the same time, when faced with the question of whether "L2 can gradually evolve into L4," Zhang Ning gave a negative answer. "The design of L4 products and autonomous driving systems is completely a different species, and the cost constraints and operational conditions of each company are also different. Although there are commonalities among companies in terms of technology, such as AI and deep learning, the success of the business depends on the top-level design from strategy to OKR to goals, which shape the corporate culture. We do not agree that L2 can gradually evolve into L4."

When discussing the commercialization and mass production timeline of Pony AI's Robotaxi models in collaboration with major automakers, Mo Luyi revealed that cooperation with partners such as BAIC and GAC is currently accelerating. "In the past, the vehicle platforms of automakers were not very suitable for autonomous driving in terms of chassis design, so the time spent was relatively long. However, with the rapid development of the Chinese automotive industry, especially as many automakers begin to venture into L3-level autonomous driving, the redundancy capability of vehicle chassis has significantly improved. In this regard, the development cycle for new models has been greatly shortened. Now, it takes about half a year to complete the full adaptation of a model, including all testing, validation, and licensing. Therefore, starting from the second half of next year, models including those recently announced by BAIC and GAC will be successively launched. Compared to three years ago, this process is undoubtedly accelerating."