SAN FRANCISCO — Uber has formed partnerships with more than 30 autonomous vehicle companies over the past two years, building a broad network that includes strategic investments in select firms—a sharp departure from its earlier push to develop self-driving technology in-house.
Uber launched its Advanced Technologies Group in 2014 under then-CEO Travis Kalanick, recruiting dozens of researchers from Carnegie Mellon University's robotics program to build proprietary self-driving technology. The effort proved expensive: research, development and testing consumed significant capital, weighing on the company's profitability targets.
The current model cuts Uber's direct R&D burden. Rather than building full-stack autonomous systems, Uber integrates solutions from specialized providers across multiple markets, accessing external innovation without the capital outlay.
The business model now centers on platform integration and network effects. Uber positions itself as the primary consumer interface for autonomous rides, sidestepping the infrastructure costs of developing its own self-driving stack.
The competitive moat shifts accordingly—from technology ownership to ecosystem orchestration. Uber's platform gives AV developers a distribution channel; in return, Uber gains access to diverse technologies and wider geographic reach.
Direct equity investments in select partners align incentives and offer upside as those companies scale.
The global approach also lets Uber deploy different AV solutions in different regions based on local partnerships, supporting broader market penetration as regulatory environments vary by market.
The model carries real risks. Uber must ensure seamless operation across multiple third-party systems—a coordination challenge that grows with each new partner added to the portfolio. Service quality and reliability are now tied directly to partner performance; an operational failure at any one provider flows back to Uber's brand and customer experience.
