The short version
- Uber executives revealed that third-party AI agents arranged only four rides and approximately five hundred food deliveries during the third quarter of 2026.
- Company leadership expressed openness to integrating with external AI tools while maintaining confidence that direct app usage remains the preferred consumer experience.
- The disclosure highlights the current friction in AI adoption for transactional services and Uber's strategy to develop internal tools to retain user engagement.
Uber executives provided a rare glimpse into the early impact of artificial intelligence agents on their platform during an earnings call, revealing that third-party AI tools facilitated only four rides in the third quarter of 2026. This disclosure marks the first time the ride-hailing giant has quantified the volume of transactions arranged by external automated systems rather than direct user interaction with its application. The data suggests that despite significant industry speculation about the potential for AI to disintermediate digital platforms, actual consumer adoption remains statistically insignificant at this stage.
The volume of activity was slightly higher within the food delivery segment, where Uber recorded approximately five hundred orders fulfilled by AI agents during the week surrounding Halloween. When viewed against the backdrop of 320 million total rides and deliveries processed by the company, these figures represent a fraction of a percent of overall business. The minimal numbers indicate that the widespread shift toward using autonomous agents to book transportation or order meals has not yet materialized in any meaningful way for mainstream users.
Industry observers have long predicted that platforms like Uber and DoorDash face existential risks from disintermediation, a process where customers bypass app interfaces entirely by delegating tasks to AI assistants such as Meta’s Muse or OpenAI’s Dots. The theory posits that as these agents become more sophisticated, they will capture the initial point of contact, potentially reducing the value of the platform brands themselves. However, Uber’s latest financial reporting underscores that this transition is still in its infancy, with significant friction points preventing seamless integration between consumer intent and automated execution.
During the earnings discussion, CEO Dara Khosrowshahi acknowledged the emerging landscape while expressing a strategic desire to remain accessible to third-party developers. He stated that the company hopes to see increased traction from AI agents and intends to facilitate easier access for these tools to use Uber’s services. This stance reflects a pragmatic approach to market evolution, recognizing that blocking external integration entirely could hinder growth if consumer behavior shifts decisively toward automated assistance.
Despite this openness, Khosrowshahi emphasized the company’s confidence in its direct-to-consumer model. He argued that Uber’s own services are designed to be the most delightful and straightforward option for users, suggesting that the native app experience holds inherent advantages over third-party intermediaries. The executive noted that it is very early in the development cycle of AI agents, and the company aims to avoid dependency on any single external player while positioning itself favorably within this evolving ecosystem.
The current state of AI agents reveals substantial hurdles before they can serve as viable replacements for direct app usage in high-frequency services like ride-hailing. Technical limitations, user trust issues, and the complexity of handling real-time variables such as traffic or restaurant availability contribute to the low adoption rates observed by Uber. These friction points protect platform operators from immediate disruption, allowing them time to adapt their strategies without facing sudden revenue erosion.
Uber is actively developing its own artificial intelligence capabilities to mitigate the risk of disintermediation and keep users within its ecosystem. Earlier this year, the company introduced a Cart Assistant for Uber Eats, an AI-driven tool that generates grocery lists based on recipe images or text prompts. By embedding AI functionality directly into its application, Uber seeks to capture the convenience benefits of automation while retaining control over the customer relationship and transaction flow.
While AI companies view ride-hailing and food delivery as primary use cases for their agents and a pathway to greater profitability, convincing platform operators to embrace disintermediation remains a significant challenge. Uber has little incentive to allow external agents to replace its direct customers if it can provide similar convenience through internal innovations. The balance between cooperation with AI developers and protection of proprietary user engagement will likely define the next phase of competition in the mobility and delivery sectors.
The disclosure of these low numbers serves as a reality check for investors and analysts who may have overestimated the immediate threat posed by autonomous agents. It highlights that technological capability does not automatically translate to market adoption, particularly in sectors where user habits are deeply entrenched. As AI tools continue to mature, Uber’s strategy will likely involve a dual approach: integrating with external systems where beneficial while aggressively enhancing its own AI features to maintain dominance.
Looking ahead, the company’s position appears stable as it navigates this transitional period. By maintaining openness to third-party integration while bolstering its internal technological offerings, Uber aims to remain relevant regardless of how consumer interaction patterns evolve. The minimal impact seen in the third quarter suggests that platform operators still hold significant leverage, but the long-term trajectory will depend on how quickly AI agents can overcome current usability barriers and gain user trust.
Sources behind this briefing
Go to the original reporting
- The Verge↗Uber only fulfilled 4 rides from AI agents last quarter
