🧠 Food for thought

Implications, context, and why it matters.

Two business models shape the robotaxi path to profit

• Pony.ai and WeRide are expanding with different business models 1.
• Pony.ai takes an asset-light approach. It supplies autonomous driving technology to partners such as Toyota and shares fleet profits 1.
• Its profit plan depends on positive unit economics in each city. Guangzhou and Shenzhen offer early examples. A roughly 70% cut in bill-of-materials (BOM) costs for its seventh-generation autonomous driving kit supports that effort 23.
• WeRide also sells autonomous vehicles, including minibuses and street sweepers. Those sales made up 52.6% of its 2025 revenue 1.
• That mix helps explain WeRide’s 30.2% gross margin and Pony.ai’s 15.7% 1.

China’s hardware base supports robotaxi expansion

• These moves also aim to spread autonomous vehicle hardware and AI systems into more markets.
• WeRide plans to deploy 200,000 vehicles with Lenovo. It says the platform can cut total cost of ownership (TCO) by 84% over a vehicle’s lifecycle, compared with its earlier HPC 2.0 platform 4.
• Chinese manufacturing scale gives companies like WeRide a cost edge over Western rivals such as Waymo, Alphabet’s self-driving vehicle company. Waymo was valued at US$126 billion in a private round and still operates mainly in the U.S. 5
• WeRide secured what it called the world’s first urban Level 4 fully driverless robotaxi commercial license outside the U.S., in Abu Dhabi. That gives it a regulatory foothold in a market where American rivals are less established 2.
• The US$147.25 billion robotaxi market projected for 2033 may favor the ecosystem that best links manufacturing, supply chains, and regulatory approval across markets 6.