Uber wants to turn its millions of drivers into a sensor grid for self-driving companies

Uber’s long-term vision extends beyond merely transporting passengers. The company aims to equip its drivers’ cars with sensors to collect real-world data beneficial for autonomous vehicle (AV) companies and potentially other firms training AI models in physical-world scenarios.

This plan was unveiled by Praveen Neppalli Naga, Uber’s chief technology officer, during an interview at JS’s StrictlyVC event in San Francisco on Thursday night. He described it as a natural progression of Uber’s AV Labs program, which was announced in late January.

Naga explained, “That is the direction we want to go eventually,” referring to fitting human drivers’ vehicles with sensors. However, he emphasized the need to first understand the sensor kits and their functionality. “There are some regulations — we have to make sure every state has [clarity on] what sensors mean, and what sharing it means.”

Currently, AV Labs operates a small, dedicated fleet of sensor-equipped cars managed by Uber, separate from its driver network. However, Uber’s ambition is much larger. With millions of drivers globally, transforming even a fraction of these cars into data-collection platforms could vastly surpass the data capabilities of any single AV company.

According to Naga, the key obstacle in AV development is not the technology but the data itself. “The bottleneck is data,” he stated. “[Companies like Waymo] need to go around and collect the data, collect different scenarios. You may be able to say: in San Francisco, ‘At this school intersection, I want some data at this time of day so I can train my models.’ The problem for all these companies is access to that data, because they don’t have the capital to deploy the cars and go collect all this information.”

Positioning itself as the data hub for the entire AV ecosystem is a strategic move for Uber, especially considering its past decision to abandon self-driving car developments, a decision co-founder Travis Kalanick has openly regretted. Many in the industry have speculated that without its own self-driving technology, Uber could become obsolete as AVs proliferate globally.

Currently, Uber partners with 25 AV companies, including Wayve in London, and is developing what Naga calls an “AV cloud.” This is a repository of labeled sensor data that partners can query to train their models. Uber plans to invest more directly in these partners, allowing them to run their models in “shadow mode” during real Uber trips, simulating AV performance without deploying the vehicles.

Naga stated, “Our goal is not to make money out of this data. We want to democratize it.”

However, given the commercial potential of Uber’s data initiatives, this stance may evolve. The company has already invested in numerous AV players, and its capability to provide proprietary training data at scale could offer it substantial influence over a sector that currently relies on Uber’s ride marketplace to connect with customers.