DoorDash, during its third-quarter earnings call, disclosed that it has only received “several hundred” orders from AI-driven third-party agents like Muse, Instinct, and Fo. This figure is notably small compared to DoorDash’s average of 3 million deliveries per day, highlighting that AI agents have yet to significantly impact DoorDash's order volume. CEO Tony Xu emphasized that while AI agent ordering is a dynamic and evolving space, the majority of customers still use the standard DoorDash app for their food deliveries.

Xu elaborated on the company’s cautious but open approach toward AI agents. DoorDash is keen on understanding agent-based ordering better and aims to build cooperative relationships with third-party agent developers rather than resist the technology’s rise. He also mentioned the possibility of new advertising opportunities within these AI platforms in the future, signaling DoorDash’s interest in adapting to how consumers might increasingly engage with artificial intelligence tools.

DoorDash has invested in its own AI-based solutions as well, rolling out several assistants to enable ordering through different interfaces. These include a text-to-order agent launched in September, a command-line interface agent released in July, and the Ask DoorDash chatbot introduced in June that allows orders via prompts and photos. In addition, DoorDash launched an Ordering Connector (MCP) designed for corporate ordering through other chatbots, reflecting the company’s broader push into AI-enhanced customer experiences.

The broader AI agent ordering landscape remains unsettled, with varied approaches from major retailers. While Amazon is actively blocking personal AI agents from accessing its content, Walmart is embracing these technologies to attract customers from competitors. DoorDash's experience illustrates that although AI ordering is still emerging and limited in impact today, companies are positioning themselves to engage with and capitalize on this evolving technology over time.