DoorDash CEO Tony Xu stated during a third-quarter earnings call that external artificial intelligence agents have generated only several hundred orders for the platform. This figure contrasts sharply with the company’s total volume of just under one billion orders in the third quarter of 2026. Xu described the current adoption rate as a very low use case for a food delivery service. He noted that these outside tools are not being used to place simple food orders directly. Instead, he sees potential for such agents in long consideration purchases where they work proactively in the background. For routine tasks like ordering food, customers tend to interact with the DoorDash interface directly rather than using third-party automation. The data suggests that current AI agents lack the utility for quick, low-stakes transactions. This indicates a gap between the capabilities of these tools and the needs of everyday consumers.
The finding highlights a practical limitation for developers building autonomous agents for retail. Users prefer direct interaction with established platforms for simple needs. This reality applies to the current state of AI technology for food delivery services. It also reflects user behaviour patterns in digital commerce.
* External agents generated only hundreds of orders versus one billion total platform orders.
* Xu identified long consideration purchases as the more viable use case for automation.
* Routine food ordering remains a direct interaction between customer and DoorDash.
