Basically no one is using AI agents for their DoorDash orders

DoorDash CEO Tony Xu said AI agents have generated only several hundred orders on the platform since they became available, compared with nearly 1 billion total orders in Q3 2026. Xu sees greater potential for agents in complex, long-consideration purchases, while saying simple food orders are generally placed directly through DoorDash.

Analysis

The signal is less about AI demand than control of the customer interface. For now, external agents appear too small to alter DoorDash’s economics; ordering food is frequent, low-consideration, and already easy inside the app. That favors DoorDash’s discovery, personalization, and direct customer relationship over an agent acting as a new intermediary. It does not prove the app experience is durable: the order count is management-reported, and usage alone says little about repeat behavior or economics.

The second-order risk is a shift in bargaining power if agents become the default interface and can compare merchants, route orders across platforms, or steer customers to direct channels. That could pressure customer acquisition, advertising value, and take rates even if DoorDash continues to provide fulfillment. Conversely, an agent that merely initiates a DoorDash order may add convenience without displacing the platform.

Near term, this is not a material earnings catalyst. Over 1–3 months, watch earnings commentary and product/API changes for evidence of agent-driven order share, incremental order frequency, or costs to acquire and serve those orders. Over 6–18 months, the strategic test is whether DoorDash can retain discovery and transaction control as agent use expands. The article alone does not establish that trend or justify changing exposure.

Key Decisions for Investors

• No trade on this signal alone; do not treat the low reported external-agent usage as a near-term earnings upgrade or as proof that AI disintermediation risk is resolved.
• Monitor DASH disclosures for agent-originated orders as a share of total orders, repeat rates, order economics, and any change in merchant or customer acquisition costs. These data are needed before sizing a thesis.
• Reassess the downside case if third-party agents gain meaningful order share while routing customers across platforms or to merchant-direct channels; that would raise risk to DoorDash’s customer relationship and monetization, even if fulfillment volumes remain intact.
• The thesis is weakened if management reports sustained agent-led order growth with DoorDash retaining the transaction and customer relationship; it is strengthened if agents capture discovery while DoorDash’s order frequency, take rate, or customer acquisition efficiency deteriorates.