Ride-hailing company Uber has disclosed that artificial intelligence agents are being used to book rides at a remarkably low rate, raising questions about the real-world adoption of agentic AI technology despite widespread industry enthusiasm. During the company's third-quarter earnings call held on Wednesday, executives revealed that only four rides were booked through an AI agent during the company's busiest week in October - the week of Halloween.

The figures for Uber's food delivery arm, Uber Eats, were slightly higher but still negligible in the context of overall order volume. Chief Financial Officer Balaji Krishnamurthy told analysts on the call that Uber processed approximately 320 million trips and orders across both Uber and Uber Eats during that same Halloween week, yet only 500 of those Uber Eats orders were initiated by an AI agent.

"It is very, very early and very limited traction that we are seeing so far. Listen, we hope that there will be more traction."

These numbers arrive at a time when the technology industry has been generating significant excitement around a new generation of AI agents - software systems designed not merely to answer questions but to carry out tasks independently on behalf of users. The contrast between the hype and the actual usage data is striking.

The AI Agent Boom and Its Limits

The past year has seen a surge of investment and interest in personal AI agent startups. Among the most talked-about is Instinct, a company founded last year that has already secured funding at a valuation of $10 billion, making it one of the most heavily backed newcomers in the space.

Meanwhile, Meta's AI assistant, Meta AI, reached the top position in Apple's U.S. App Store last month, surpassing 10 million users according to figures reported by Meta itself. The company noted that a significant portion of those users were engaging with the assistant to shop and conduct transactions with businesses.

Major technology companies including OpenAI, Anthropic, Google, and xAI - the artificial intelligence company founded by Elon Musk and developer of the Grok assistant - have all launched their own AI agent products. A wide range of smaller startups have followed suit, contributing to what has become a crowded and fast-moving market segment.

Early Public Data Tells a Cautious Story

However, the limited data emerging from publicly traded companies suggests that the so-called agentic web remains in its earliest stages. DoorDash, the food delivery platform, disclosed on its own earnings call last week that it has received only a few hundred orders through AI agents to date. The vast majority of its customers continue to place orders through the DoorDash app directly, without any AI intermediary involved.

Uber appears to be experiencing a nearly identical pattern. Khosrowshahi offered some context for why AI agents may not be gaining traction as quickly as some in the industry might have anticipated. He pointed out that booking a ride through the Uber app takes an average of just 50 seconds, while re-ordering a previously saved food item takes roughly 90 seconds. Given those timeframes, it is not immediately clear that delegating those tasks to an AI agent would offer users any meaningful time savings.

Uber Remains Optimistic Despite Low Numbers

Despite the underwhelming early figures, Khosrowshahi expressed a broadly positive view of AI agents as a long-term opportunity for Uber's business. He framed the technology as an additional channel through which the company could attract demand rather than a replacement for existing booking methods.

"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 also noted that there are currently around eight significant AI assistant platforms operating in the market and suggested that figure is likely to grow beyond ten over time. He indicated that Uber is in active conversations with companies across the AI landscape, adding that every major AI systems company has approached Uber about potential integrations.

The data from Uber and DoorDash provides a grounded counterpoint to the broader narrative surrounding AI agents, illustrating that consumer behavior tends to evolve more gradually than technology industry timelines often predict. While the infrastructure for AI-driven commerce is clearly being built, actual user adoption remains at an early and modest stage.