🔍 Read the full analysis: DoorDash’s 993 Million Deliveries Put AI Shopping Claims In Context on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
DoorDash CEO Tony Xu said third-party AI agents have generated “maybe several hundred orders in total” since they began operating with the delivery platform. DoorDash handled 993 million orders in the third quarter of 2026, putting the agent figure in sharp scale context, though Xu’s estimate covers an unspecified period and is not a quarterly count.
DoorDash CEO Tony Xu said third-party AI shopping agents have generated “maybe several hundred orders in total” since they began operating, a small volume compared with the 993 million orders DoorDash handled in the third quarter of 2026. The remarks, made during the company’s October 29 earnings call, offer a measure of how limited outside agents’ ordering activity remains on a major delivery platform, despite rising claims about AI-assisted shopping.
Xu gave the estimate in response to an analyst asking whether DoorDash would power food-ordering agents itself or leave that work to general-purpose agents such as Meta’s Muse. He described the area as “still a very dynamic” one, but said that for an easy task such as ordering food, the predominant use remains through DoorDash’s own app. Later in the call, Xu repeated that agents had generated several hundred orders across the time they had been available and called it “a fairly low use case.”
DoorDash’s third-quarter results, released October 29, reported 993 million total orders, up 28% from a year earlier. The agent estimate is rounded and refers to the entire time the agents have been operating, not just the quarter. If “several hundred” is treated generously as 1,000 orders, and all are assumed to have fallen in the quarter, that would be roughly one order per million DoorDash orders. That is an illustrative calculation, not a company-reported share or directly comparable quarterly measure.
Xu also said that advertising opportunities could emerge within AI agents, but that low current volumes mean “there isn’t yet that value.” DoorDash’s call therefore offered both a usage estimate and a qualification: the company sees a potential business opportunity, but said present activity is too limited to establish its value.
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
Agent Orders Remain Tiny at DoorDash
The comparison matters because food delivery is a familiar, repeatable purchase that may seem well suited to an assistant: users can specify a meal, location and budget, then place an order. Yet Xu’s remarks suggest that, at least on DoorDash, third-party agents have not displaced the company’s direct ordering channel at meaningful scale. The estimate does not show that consumers never use AI for shopping; it describes orders attributed to outside agents on one platform.
For delivery companies, the route through which customers place an order can affect control over the customer relationship, advertising and other revenue. Xu’s comments indicate that DoorDash is watching the possibility of agents becoming an additional channel, while saying current order volumes are too low to support substantial advertising value. For customers, the figures temper broad claims that AI agents are already handling routine purchases at scale.
Products Worth Considering
Amazon Kindle Paperwhite (12th Generation, 2024), 16 GB, 7-inch Display, Jade
Sonos Era 100 SL Compact, Microphone-Free Speaker – Black
Bose TV Speaker Compact Soundbar with Bluetooth and HDMI-ARC, Black
As an affiliate, we earn on qualifying purchases.
Earnings Calls Offer a Reality Check
DoorDash was one of several large companies discussing agent-led shopping in late October. On October 29, Amazon CEO Andy Jassy said “few customers are doing much shopping this way yet,” adding that the experience was not yet better or more convenient than shopping directly with a retailer. Visa CEO Ryan McInerney, speaking October 27, described the number of agentic transactions as limited and called trust a major barrier to scale.
Meta has presented a more upbeat picture of its Muse personal agent. CEO Mark Zuckerberg said on October 21 that 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 said more than a quarter of users engaged in shopping and that the average purchase size exceeded $200. Those are Meta’s own figures; the source material provides no independently audited sales total, and a reported average purchase size does not establish how many transactions were completed or how representative they were.
The companies’ measures are not directly interchangeable: DoorDash discussed orders coming from outside agents, while Meta described Muse users and shopping activity. The claims point to different aspects of adoption, and neither alone establishes the overall amount of commerce being completed through AI agents.
““We’re seeing maybe several hundred orders in total from some of these third-party agents.””
— Tony Xu, DoorDash CEO, on the October 29 earnings call
The Agent Order Count Has Limits
DoorDash did not provide an exact number, a start date for the period covered, or a breakdown by agent, order type or customer. Xu’s “several hundred” is a rounded spoken estimate covering the agents’ full operating period, so it cannot be treated as a precise third-quarter total. The one-in-a-million illustration relies on an assumed 1,000 orders and assumes all occurred during the quarter.
It is also unclear how much shopping through agents is taking place outside DoorDash, how often people use agents to research or compare products without completing a purchase, and whether orders are being counted consistently across companies. Meta’s reported user and purchase figures do not resolve those questions, and its statements are company-reported rather than independently audited in the source material.
Watch for Orders and Repeat Use
DoorDash did not announce a timetable for expanding third-party agent ordering or report a revenue-sharing agreement with an agent provider. Xu said the company sees potential advertising opportunities in agent environments, but described current volumes as too low for that value to be present yet. The next useful evidence would be more specific order totals, a defined measurement period and information about repeat use.
Future earnings calls and company disclosures may show whether agent shopping grows beyond early use and whether customers find it more convenient than ordering directly. Until companies provide comparable transaction data, claims about adoption should be read with attention to what each figure measures: users, shopping activity, completed purchases or orders on a particular platform.
Key Questions
How many orders did DoorDash say came from outside AI agents?
CEO Tony Xu said the agents had generated “maybe several hundred orders in total.” He did not give an exact count or specify a quarterly total.
How many orders did DoorDash handle in Q3 2026?
DoorDash reported 993 million orders in the third quarter of 2026, up 28% from the same quarter a year earlier.
Does the comparison mean AI agents handled one in a million DoorDash orders?
Not as a reported company statistic. One in a million is an illustration based on assuming 1,000 agent orders and placing all of them in the quarter. Xu’s estimate covers the agents’ entire operating period, whose duration was not specified.
What did DoorDash say about making money from agent shopping?
Xu said advertising opportunities could arise within AI agents, but that low current order volumes meant there was not yet value from that opportunity.
Does Meta’s Muse data contradict DoorDash’s figures?
Not directly. Meta reported user and shopping activity for Muse, while DoorDash discussed orders placed through outside agents on its platform. The figures measure different things and do not establish a comparable total for agent-led commerce.
Source: ThorstenMeyerAI.com
