Pony.ai, Inc. Q4 FY2024 earnings call

Management Statement and Operational Highlights

• Technological Readiness: Achieved level 4 driverless operation 24/7 in all weather conditions. PonyWorld simulation enables virtual driver training, with the virtual driver learning through billions of generative simulation trials to handle complex scenarios.
• Regulatory Support: Secured fully driverless fair charging licenses in China's Tier 1 cities, with supportive regulatory environments in these cities.
• Mass Production Partnerships: Joint ventures with Toyota, BAIC New Energy, and GAC Group for robotaxi production. Partnerships include co-development of vehicles and value chain integration (maintenance, charging).
• Operational Capabilities: Developed Pony Pilot ride-hailing platform and partnerships with TNCs like Amap, Alipay to expand service accessibility.
• Robotruck Business: Transformed joint venture with Sinotrans into an autonomous transportation platform, and approved for Robotruck driver-out platooning across provincial highways.

Segment Performance

• Robotaxi services: 2024 revenue was US$7.3 million, down 5.3% year-over-year (YoY). Passenger fares saw growth due to expansion of public-facing fair charging robotaxi operations in Tier 1 cities.
• Robotruck services: 2024 revenue was US$40.4 million, up 61.3% YoY, driven by fleet expansion into new regions.
• Licensing and application revenue: 2024 revenue was US$27.4 million, down 30.1% YoY, influenced by recognition schedule of project-based revenue.
• Total revenue: Full year 2024 totaled US$75 million, up 4.3% YoY. Total cost of revenue was US$63.6 million, up 15.6% YoY. Gross profit was US$11.4 million, with a gross margin of 15.2% (down from 23% in 2023).

Guidance

• Continued volatility in quarterly revenue and margins is expected due to the nascent stage of commercialization.
• Confident in scaling up commercialization, driving sustainable growth, but no formal guidance provided at this time.

Risks

• Volatility in revenue and margins due to early commercialization stage.
• Risks associated with forward-looking statements, where actual results may differ from expectations.
• Challenges in maintaining safety and cost efficiency as fleet scale increases.

Question and Answer

Q: What's the strategic rationale behind your robotaxi first, China first, and also the Tier 1 cities first approach?

A: James Peng stated China has the largest ride-hailing market with supportive regulations and growing user demand. Tier 1 cities in China are ideal for mass deployment due to technical challenges and regulatory clarity.

Q: Could you differentiate your business model against the OEM ride sharing company and also the taxi company and any collaboration with these companies?

A: Leo Wang explained Robotaxi service uses a virtual driver, OEMs sell co-developed vehicles, and ride-hailing platforms use virtual drivers as part of their driver pool, creating a win-win value chain.

Q: Do you foresee any challenges before mass commercialization?

A: James Peng stated technology, regulation, mass production, and large-scale deployment are in place; confident in no insurmountable challenges.

Q: How do you achieve the very high safety level compared to human driver and why you believe the level four autonomous driving technology depends on more?

A: Tiancheng Lou explained L4 uses reinforcement learning in a virtual world, not relying on massive real-world data, as imitation learning is limited by human performance ceilings.

Q: Why were 2024 costs and expenses higher year-over-year and any guidance on costs going forward?

A: Leo Wang mentioned increased R&D for seventh-generation vehicle development, with continued expenditure expected as the focus is on deploying these vehicles in Tier 1 cities.

Q: How do disruptive technologies like DeepSeek play out in L4 autonomy development?

A: Tiancheng Lou said disruptive tech can impact aspects but singular breakthroughs provide marginal help; success depends on multiple factors like safety, cost, and partnerships.

Q: Details on cooperation with OEMs and progress?

A: James Peng discussed joint ventures with Toyota, BAIC, and GAC to reduce unit costs and enable mass production of robotaxis.