The effort aims to create an “AV cloud”—a library of labeled sensor data that partner firms can query to train models, potentially enabling shadow mode testing on real Uber trips.

Regulatory considerations are acknowledged, focusing on sensor regulations and data-sharing rules across different states and regions.

The initiative targets data bottlenecks in AI training by providing diverse, high-quality physical-world data for AV models, including niche scenarios and diverse environments.

Uber must navigate hardware costs and varied data privacy and sensor-use regulations across regions before full deployment.

Uber plans to outfit its human drivers’ cars with sensors to collect real‑world data for autonomous vehicle companies, expanding its AV Labs initiative and helping build an AV data backbone.

With 25 AV partnerships and ongoing investments in autonomous players, Uber could leverage large-scale data to gain influence in the evolving AV ecosystem.

According to Neppalli Naga, data is the bottleneck for AV development, and Uber’s data could unlock access to diverse driving scenarios and locations for a wide range of companies.

Uber’s motivation emphasizes democratizing access to data rather than direct monetization, though the move could shift competitive dynamics in the AV industry.

The program positions Uber as an essential infrastructure layer for the self-driving ecosystem by prioritizing data accessibility over immediate profits.

Sensor kit performance is being tested across environments to assess feasibility and impact as Uber expands its AV investments beyond its own self-driving car unit.

Overall, the initiative aims to keep Uber relevant in a driverless future by using its rider network as a data backbone for AV developers.

CTO Praveen Neppalli Naga announced the initiative, aiming to scale from a small AV Labs program to a distributed data-collection platform using independent contractors worldwide.