Uber’s long-term vision extends beyond simply transporting passengers. The company aims to equip its drivers’ cars with sensors to gather real-world data for use by autonomous vehicle (AV) and AI companies. Praveen Neppalli Naga, Uber’s CTO, discussed this plan at TechCrunch’s StrictlyVC event in San Francisco, highlighting it as a continuation of their AV Labs initiative announced in January.
Naga mentioned that while the ultimate goal is outfitting drivers’ cars, initial steps involve understanding sensor kits and navigating regulatory challenges. Currently, AV Labs utilizes a small fleet of Uber-operated, sensor-equipped cars, separate from the driver network. However, the broader vision is to leverage Uber’s extensive global driver network for large-scale data collection, providing a significant resource for the AV industry.
According to Naga, data is the current bottleneck in AV development, not technology. Companies like Waymo need varied real-world data for model training, which Uber’s infrastructure could supply. Uber, having shifted focus from creating its own self-driving cars, now seeks to become a crucial data provider for the AV sector. The company, which partners with 25 AV enterprises, is developing an “AV cloud” for sharing labeled sensor data to train partner models.
Uber has emphasized that its intention is not to profit directly from this data but to democratize it. However, given the strategic advantage this initiative provides, this stance might evolve. Uber’s existing equity investments in various AV companies and its ability to deliver unique training data could offer significant advantage within the industry, which relies heavily on Uber’s ride-sharing platform.
Uber has already made investments in AV players, and its data-centric approach could provide substantial leverage over a sector reliant on its marketplace for customer access.
