Uber is pursuing a future where its drivers serve a purpose beyond merely transporting passengers and delivering food. The company has disclosed its intentions to harness the vehicles used by its drivers as mobile data collectors. This initiative aims to collect real-world information for companies developing self-driving technologies and artificial intelligence models, as revealed by Uber’s Chief Technology Officer, Praveen Neppalli Naga, during an interview at TechCrunch’s StrictlyVC event in San Francisco.

Naga emphasised that while the vision is ambitious, there are preliminary steps that need to be addressed, including understanding sensor systems and establishing regulations surrounding privacy and data sharing. Currently, Uber operates its Autonomous Vehicle (AV) Labs programme, which relies on a limited fleet of company-managed cars equipped with advanced sensors. These vehicles are distinct from the broader Uber driver network, but the ultimate goal is considerably larger.

With millions of drivers globally, even a fraction of them utilising sensor-enabled cars could enable Uber to establish one of the most extensive real-world driving data networks in existence.

Challenges and Opportunities in Data Collection

Naga indicated that the significant challenge for companies working on autonomous vehicles has shifted from developing technology to acquiring substantial and useful data for training self-driving systems. He pointed out that specific data is often necessary, such as videos from school crossings at particular times or traffic patterns at busy intersections. Obtaining this information typically requires substantial financial resources and large fleets of vehicles, which many companies lack.

Uber views this gap as an opportunity. Instead of assembling its own fleet of autonomous taxis from the ground up, the company envisions itself as a key data provider for the broader self-driving sector. To date, Uber has formed partnerships with approximately 25 autonomous vehicle firms, including Wayve, based in London.

In addition to its partnerships, Uber is developing what Naga refers to as an “AV cloud,” which will serve as a comprehensive database of labelled sensor data available to partner organisations for training their AI systems. Furthermore, Uber is enabling its partners to test their artificial intelligence in what is called “shadow mode.” This approach allows a trained self-driving model to run virtually during an actual Uber trip, providing insights without deploying an autonomous vehicle on public roads.

Impact of AI on Uber’s Operations

In parallel with its ambitions to support external self-driving AI, Uber is undergoing significant internal transformations driven by AI technology. Naga noted that the company’s initial budget projections for AI-related projects have already been exceeded due to the rapid adoption of sophisticated coding tools such as Anthropic’s Claude Code.

He explained that the swift escalation in AI-related expenditures has prompted him to reassess budgetary needs, illustrating the fast-paced nature of technological advancements. Internally, Uber asserts that software development processes are undergoing deep alterations as AI tools evolve from merely assisting with suggestions and auto-completion to taking on more autonomous coding roles.

Naga categorised this progression as “agentic software engineering,” where AI systems autonomously generate code and conduct tasks with minimal human intervention. This shift is already visible, with Uber reporting that approximately 1,800 code changes are now generated entirely by its internal AI coding agent each week. Notably, nearly 95 per cent of Uber engineers utilise AI tools monthly, and close to 70 per cent of the code being committed now originates from AI-assisted systems, marking a shift from less than 1 per cent to around 8 per cent within a few months.

Long or Short, get news the way you like. No ads. No redirections. Download Newspin and Stay Alert, The CSR Journal Mobile app, for fast, crisp, clean updates!

App Store –  https://apps.apple.com/in/app/newspin/id6746449540 

Google Play Store – https://play.google.com/store/apps/details?id=com.inventifweb.newspin&pcampaignid=web_share