Uber Reports Minimal AI Agent Ride Bookings Despite Growing Industry Hype

Commenting on the metrics, Uber Chief Executive Officer Dara Khosrowshahi emphasized how early the ecosystem remains. "It is very, very early and very limited traction that we are seeing so far," he said regarding the agent booking numbers, adding, "Listen, we hope that there will be more traction."

Efficiency challenges and industry adoption

Uber's experience mirrors trends seen elsewhere in the on-demand economy. Food delivery platform DoorDash disclosed on its earnings call last week that it has received only a few hundred orders initiated by AI agents, as the vast majority of consumers continue placing orders directly through its native app.

Part of the reason for this low engagement lies in the convenience of established user interfaces. Khosrowshahi highlighted that ordering a ride through Uber's standard mobile application takes an average of only 50 seconds. Similarly, re-ordering a favorite meal on Uber Eats requires just 90 seconds. With existing digital workflows already optimized for speed, delegating simple tasks to an AI assistant provides little practical time savings for end users.

These early consumer adoption metrics stand in sharp contrast to the aggressive capital investment and user growth seen across the AI sector. Major platforms including OpenAI, Anthropic, Google, and SpaceXAI (Grok) are actively offering agentic capabilities. High-profile startups are also attracting large investments, such as Instinct, which was founded last year and secured its latest funding round at a $10 billion valuation. Additionally, Meta reported that its Muse app reached number one in the Apple U.S. App Store last month, topping 10 million users who used it for shopping and business interactions.

What it means for developers

For developers building next-generation AI agents and tool-use integrations, Uber's earnings data offers an important reality check. While agentic frameworks are technically capable of booking services via APIs or web automation, user adoption depends heavily on whether an agent solves real friction. In scenarios where native mobile apps already complete a task in under a minute, consumers have little incentive to switch to a conversational interface.

As engineers refine function calling, browser automation, and API orchestrations, evaluating model capabilities across different providers is essential. Developers can try top AI models cheaply through one API at https://apixoai.online to test how different LLMs perform on agentic execution and task planning.

Looking forward, Uber views AI assistants as an additional channel for platform growth rather than a disruption. "We think that is another opportunity to reach our platform. We think it will be a net positive in terms of demand on the ecosystem," Khosrowshahi noted. He observed that there are currently about eight major AI assistants on the market, expecting that number to grow past ten, and confirmed that "every single AI system company is talking to us."

Source: Almost no one uses AI agents to book their Uber, company says — TechCrunch AI. Written by the Apixo team from that report.