The Four-Ride Reality Check
During Halloween week, one of its busiest periods, Uber processed roughly 320 million trips and deliveries. Amidst this massive volume, a surprisingly tiny number emerged from the company's Q3 earnings call: just four rides were booked by external AI
agents. The food delivery side, Uber Eats, saw slightly more but still minimal activity, with around 500 orders placed by these agents. Uber's CEO Dara Khosrowshahi acknowledged the figures, stating, 'It is very, very early and very limited traction that we are seeing so far'. This stark data provides a crucial reality check on the current state of 'agentic commerce', where autonomous AI is supposed to handle our digital errands.
What Are AI Agents Anyway?
Unlike the voice assistants we're used to, such as Siri or Alexa, which primarily respond to questions, AI agents are designed to take action. An AI agent is a piece of software that can understand a complex goal—like 'plan a weekend trip to Goa' or 'order a healthy dinner for four'—and then execute the necessary steps across different apps and websites to achieve it. This involves not just searching for information but also making choices, filling out forms, and completing payments on a user's behalf. Companies like Meta, Google, and OpenAI are all developing these agents, promising a future where you can delegate your digital to-do list to an intelligent, autonomous assistant.
The Challenge of Simple Tasks
So why the low numbers? One primary reason is that for many services, the existing process is already incredibly efficient. Uber's leadership noted that booking a ride directly in their app takes an average of only 50 seconds. For a task that is already so quick and simple, introducing an AI agent as a middleman doesn't offer a significant advantage and may even add complexity. This suggests that AI agents will likely gain traction first on more complex, multi-step tasks rather than simple, quick transactions that are already optimized.
DoorDash and Airbnb See a Similar Trend
Uber isn't alone. DoorDash CEO Tony Xu reported that third-party agents had generated only 'several hundred orders in total' since their launch, a tiny fraction of the platform's volume. Similarly, Airbnb CEO Brian Chesky described the number of bookings coming from external AI agents as 'very non-material'. Chesky suggested that travel planning is an experience many people enjoy, involving visual browsing and social decision-making that a simple chatbot can't replicate. This has prompted these companies to build their own specialized AI tools. DoorDash has launched 'Ask DoorDash', a conversational tool to build grocery carts from recipes or text prompts. Airbnb plans to roll out its own sophisticated AI travel agent in 2027, designed to help users plan entire trips with a rich, visual interface.
The Future of Autonomous Commerce
Despite the slow start for third-party agents, none of these companies are dismissing AI. Instead, they see it as a powerful tool to be integrated into their own platforms. Uber is developing its own in-app voice agents and cart assistants. The consensus is that while users may not be ready to hand over their credit cards to a general-purpose external AI, they are open to AI-powered features that make existing apps smarter and more helpful. The four Uber rides are not a sign of failure, but rather an early data point in the long and complex integration of AI into our daily commerce. The platforms that own the customer relationship, like Uber, DoorDash, and Airbnb, are positioning themselves to set the terms for how—and where—this AI-driven future unfolds.
















