🔍 Read the full analysis: A Reality Check On AI Shopping Agents, Using DoorDash’s Order Data on ThorstenMeyerAI.com
Get business pricing on office and shipping supplies
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
DoorDash CEO Tony Xu said third-party AI agents have generated “maybe several hundred orders in total” since they began operating on the platform. DoorDash handled 993 million orders in the third quarter of 2026, though the agent figure covers an unspecified period and is not a quarterly count. Amazon and Visa executives also described agent shopping as limited, while Meta has reported stronger shopping activity among users of its Muse agent.
DoorDash Chief Executive Tony Xu said on October 29 that third-party AI agents had generated “maybe several hundred orders in total” since they began operating on the platform, a small figure beside the company’s 993 million orders in the third quarter of 2026. The comments offer a concrete, though limited, measure of how little outside AI agents are currently being used to place orders through one major delivery service.
Xu gave the figure during DoorDash’s third-quarter earnings call, responding to an analyst who asked whether the company wanted to power food-ordering agents itself or leave that role to general-purpose assistants such as Meta’s Muse. Xu called the area “still a very dynamic” one, but said that for a relatively simple task such as ordering food, the predominant use case remained DoorDash’s own app.
Xu repeated the estimate later in the call, describing agent orders as “several hundred” across the time the agents had been operating and calling the use case “fairly low.” He also said advertising opportunities might eventually emerge inside AI agents, but that the low order volumes meant there was not yet value to share. DoorDash’s earnings release reported 993 million total orders in the quarter, up 28% from a year earlier.
The figures do not cover matching periods: Xu’s agent estimate is cumulative over an unspecified operating period, while 993 million is the quarter’s total. Treating “several hundred” as 1,000 and, generously, assigning all those orders to the quarter would put them at roughly one in a million of DoorDash’s quarterly orders. That is an illustrative calculation, not a company-reported rate.
Infographic 01 · The number
993 million orders. “Several hundred” from AI agents.
What DoorDash’s CEO said about orders placed by outside AI agents, next to the company’s own order count.
Drawn to scale
Q3 2026 orders vs. a generous upper bound for agent orders
The second bar is not missing. At this scale it is about one ten-thousandth of one percent wide.
“You never want to fight gravity. But at the same time, you don’t want to be delusional … we’re seeing maybe several hundred orders in total from some of these third-party agents.”Tony Xu, CEO of DoorDash · Q3 2026 earnings call, 29 Oct 2026
Sources: DoorDash Q3 2026 results (29 Oct 2026); DoorDash earnings call (29 Oct). Ratio and per-day figure: my calculation. “Several hundred” is a spoken, rounded figure.
Powered by Thorsten Meyer AI
Why Agent Orders Remain Small
The comparison suggests that, at least on DoorDash, AI shopping interest has not yet translated into meaningful order volume through outside agents. For shoppers, the immediate implication is that a conversational assistant may help with discovery or recommendations without becoming a common substitute for ordering directly through a retailer or delivery app.
For platforms and advertisers, scale matters. Xu said DoorDash does not yet see enough volume to create value from advertising within third-party agents. If agents do become a regular route to purchase, companies may have to decide how to handle payments, customer relationships and advertising attribution. The call did not establish how those arrangements would work.
The DoorDash number is not a measure of all AI-assisted shopping. It covers orders attributed to third-party agents on this service, not people who use an AI assistant to research a product and then buy through an ordinary app or website. It also cannot settle whether agent shopping will grow: it records current activity, not a forecast.
Popular right now
Samsung 85-Inch QN70H Neo QLED 4K Smart TV (2026)
Ninja DualZone 6-in-1 Air Fryer, 8-Qt, Grey, DZ801
Dyson Airwrap i.d. Multi-Styler for Straight and Wavy Hair
As an affiliate, we earn on qualifying purchases.
Executives Describe Early Adoption
DoorDash’s comments came amid a series of mixed signals from large technology and payments companies. On October 29, Amazon CEO Andy Jassy said few customers were doing much shopping through outside, general-purpose agents, arguing that the experience was not yet better or more convenient than shopping directly. That assessment, like Xu’s, was made on an earnings call and reflects the company executive’s account.
Visa CEO Ryan McInerney said on October 27 that agentic transactions remained limited and that trust was the biggest barrier to scale. These remarks point to a challenge beyond whether an agent can search for an item: customers must also be comfortable authorizing an automated system to make a purchase and use payment credentials.
Meta has described more active shopping use of its Muse personal agent. On October 21, Mark Zuckerberg said more than 10 million people in the United States and Canada had set up Muse in about six weeks, and that shopping was its most common use. Meta also cited an average purchase above $200. Those are Meta’s own figures; the source material provides no total sales value, and the metric does not show how frequently users make everyday purchases such as meals.
““We’re seeing maybe several hundred orders in total from some of these third-party agents.””
— Tony Xu, DoorDash CEO, on the company’s October 29 earnings call
Limits of the Order Estimate
DoorDash did not specify how long the “several hundred” orders accumulated over, which third-party agents were included, or how the company identifies and counts those orders. The estimate is rounded, and the company did not provide a detailed breakdown by agent or type of purchase. That makes a direct comparison with quarterly orders illustrative rather than a like-for-like adoption rate.
It is also unclear whether low agent-attributed order counts reflect limited customer interest, restrictions on how agents can interact with services, payment and trust concerns, or some combination. Meta’s reported shopping activity uses different measures and does not establish how many purchases were completed through its agent. The available statements do not provide a common industry-wide measure of AI-driven sales.
What DoorDash Will Watch
DoorDash has not announced a timetable for when third-party agent orders might become a material part of its business. Xu said advertising within agents could present an opportunity later, but described current volumes as too low to create value. The next useful evidence would be a comparable order count or share of sales, with a stated measurement period and clear definition of which transactions qualify as agent-driven.
Further earnings calls and company disclosures may show whether customer use grows and whether platforms open more access to outside agents. For now, DoorDash’s reported figure provides a snapshot of activity on its service, while the longer-term effects on shopping, delivery and advertising remain unsettled.
Key Questions
How many orders did AI agents generate on DoorDash?
CEO Tony Xu said third-party agents had generated “maybe several hundred orders in total” over the period they had been operating. DoorDash did not state the exact count or period.
How does that compare with DoorDash’s order volume?
DoorDash reported 993 million orders in the third quarter of 2026. The agent estimate covers an unspecified cumulative period, so it cannot be directly compared with the quarter. A calculation assuming 1,000 agent orders all fell within that quarter would be roughly one in a million, but that is only an illustration.
Does the DoorDash figure cover all AI-assisted shopping?
No. It refers to orders attributed to third-party agents on DoorDash. It does not count every shopper who uses AI for research before completing a purchase through a regular app or website.
Why might customers not use agents to place more orders?
The executives cited in the source material described the experience as not yet more convenient than ordering directly and identified trust as a barrier. DoorDash did not provide data explaining why its agent order count is low.
Source: ThorstenMeyerAI.com
