Uber uses OpenAI to help people earn smarter and book faster

Uber is leveraging OpenAI frontier models to power new AI-driven features across its global marketplace, according to OpenAI. Operating at a scale of 40 million trips per day across more than 70 countries, Uber is using large language models to reason across complex real-time signals, deliver conversational responses, and build new voice experiences within its app. For drivers and couriers, Uber has developed Uber Assistant, an AI-powered tool designed to help users navigate marketplace data, earnings trends, and positioning insights. The assistant uses a multi-agent architecture that routes different tasks to specialized models, ranging from lightweight nano and mini models for fast responses to larger reasoning models for complex tasks. To maintain safety and privacy, Uber built AI Guard, an internal governance layer that screens prompts and responses, enforces policies, and reduces hallucinations. In addition to supporting drivers, Uber is applying OpenAI Realtime APIs to introduce voice booking features for riders. The system interprets natural speech intent, accounts for saved locations and context, and synchronizes spoken and visual responses in the application. Voice functionality is designed to remove multi-tap barriers and improve accessibility for travelers with heavy luggage, older adults, and visually impaired users. The collaboration has also changed how internal teams operate, allowing product, legal, operations, and design groups to collaborate more closely on prompting, evaluation pipelines, and orchestration frameworks rather than relying solely on a centralized AI team. Uber Assistant has been expanded in an experimental rollout to hundreds of thousands of U.S. drivers, showing strong repeat engagement and helping new drivers accelerate their onboarding and platform ramp-up.