Shopify AI transaction orders have increased 13-fold.
Shopify AI transaction orders have increased 13-fold.
On the morning of October 16, pre-orders began for Apple’s first foldable smartphone, the iPhone Duo. Priced from $1,999, initial production capacity was tight, and delivery dates were soon pushed back to January next year.
This time, the rush to be first isn’t limited to those glued to their screens.
This year, Agent has become the harshest parent to iPhone Duo scalpers | Image source: Apple
On the MacRumors forum, a user who failed to secure one complained bitterly, while on Reddit, some people were bragging that their Muse bot had snatched up several units in one go—before any real users could even click, the inventory was already swept away by AI.
Muse is the personal agent launched by Meta on September 8. Upon its release, it briefly topped the free charts in the U.S. App Store, with downloads quickly surpassing 2.5 million. According to Meta, shopping has become one of Muse’s primary use cases. Just around the time the iPhone Duo went on sale, Muse even released a new version optimized for this new device; the accompanying update notes prominently highlighted making in-chat shopping more convenient.
AI-powered shopping on behalf of users has moved from demo presentations at launch events to real-world rush‑buying scenarios.
But not every store welcomes these new customers. In late September, Amazon began banning Muse. By October, users of Musk’s Grok Bot encountered the same warning when placing orders on Amazon: unauthorized AI agents continue to access the platform, violating Amazon’s terms of service.
Another company, Shopify, took the exact opposite approach. On September 21, Shopify CEO Tobi Lütke announced a deep partnership with Muse, enabling Muse to use Shop Pay to complete checkout across all Shopify stores.
Meanwhile, Shopify’s financial report has already demonstrated the early returns of this approach. In the first quarter of this year, orders generated by AI-powered search increased nearly 13-fold year over year, while store traffic driven by AI grew eightfold.
Closing one door while opening another. As more and more purchasing decisions are handed over to Agents, could “whether or not to let the Agent in” become a key factor in determining e-commerce rankings in the AI era?
The space behind 13x growth
Shopify is the most active “player” in this round of the AI shopping boom.
In the fourth quarter of 2025, it reported a 15-fold increase in AI-driven orders. In the first quarter of 2026, store traffic driven by AI grew 8-fold year over year, while orders generated through AI search increased nearly 13-fold—and new buyer orders from AI search grew at almost twice the rate of those from traditional organic search. In the same quarter, Shopify posted its fastest revenue growth in more than four years.
By the second quarter, the multiple had fallen back to 3x, with AI-driven traffic and orders both tripling compared to the same period last year. However, this quarter’s structure is even more compelling: 75% of AI‑attributed purchases occurred outside the platform’s top 100 categories, and AI searches conducted via Shopify Catalog delivered a conversion rate twice that of standard AI searches.
This set of data indicates “rapid growth,” not “large scale.”
Shopify has been quite candid about this. During the second-quarter earnings call, CEO Harley Finkelstein acknowledged that agentic commerce transaction volume remains very small compared to the platform’s massive GMV. To date, Shopify has not disclosed either the absolute value or the share of GMV generated through AI channels, while its total GMV for the quarter was $116 billion.
Muse Agent wallet page supports Shopify’s Shop Pay | Image source: GeekPark
From 15x to 13x and then to 3x, the multiples are converging rapidly—this is a typical pattern as the low-base effect fades. The scope is also shifting: in the first quarter, the focus was on “AI search orders,” while in the second quarter it shifted to “AI‑driven traffic and orders.” As a result, the two sets of figures may not align into a single, continuous trend line.
Shopify’s second-quarter revenue grew 34% year over year, reaching $3.6 billion. Driving this growth were a 32% increase in GMV, an uptick in Shopify Payments adoption to 68%, and a 53% year-over-year rise in Shop Pay GMV. Finkelstein also stated clearly that transactions generated through Agent are charged at the same rate as those from regular online stores, with no additional fees. In other words, the revenue contribution of Agent orders is embedded within GMV and payment metrics and cannot be isolated separately.
But the significance of 13x doesn’t lie in how much revenue it generates today; rather, it demonstrates that Agent can bring Shopify customers that it previously couldn’t reach.
The first signal is new buyers. The growth rate of new customer orders on AI-powered channels is about twice that of other channels, and Shopify specifically emphasizes that traditional search remains one of merchants’ largest traffic sources—and is still growing. In other words, AI isn’t just redirecting traffic that originally came from search; it’s also driving incremental growth.
The second signal is the long tail. Seventy-five percent of AI orders fall outside the top-tier categories, which happens to be Shopify’s core strength. Millions of independent small stores have historically struggled to gain visibility in the competitive landscape of search ads and e‑commerce platforms, but Agent excels at matching products to specific needs, giving these small shops their first chance to be “found.”
To help merchants see this, Shopify added an agentic section to its backend in May this year, allowing merchants to view orders and conversions from ChatGPT, Copilot, Google AI Mode, Gemini, and its own Shop App. What Shopify is doing is turning the Agent—from a concept—into a visible sales channel within the merchant’s backend.
Amazon isn’t blocking the Agent,
It’s someone else’s Agent
Amazon’s tough stance toward external agents has a long history.
In November 2025, Amazon sued Perplexity, alleging that its shopping agent in the Comet browser accessed Amazon’s systems without authorization. In March this year, the court granted a preliminary injunction, and Perplexity has since appealed to the Ninth Circuit Court of Appeals.
Muse is the second one. Amazon’s reasons were very specific: Meta never informed users that Muse would access the Amazon store, Muse did not identify itself during browsing, and it allegedly scraped and stored users’ account credentials. Amazon had asked Meta to proactively remove Amazon from Muse, but when Meta refused, Amazon simply banned it outright.
Grok Bot is very likely the third one. Amazon has not publicly confirmed this, but several users have already shared the same blocking message. Musk first advised users to have their bots shop elsewhere, even showing his own bot successfully placing an order on LEGO’s official website, before jokingly adding, “Maybe we should build a competitor to Amazon.”
Security concerns are not unfounded. Recently, a Canadian user publicly complained that Muse, without his consent, accepted a buyer’s offer on the Marketplace and even shared his home address with the buyer. With an agent operating under the user’s account, no clear answer has yet emerged as to who would be held responsible for any mistakes.
But security is only the reason on the surface; beneath the surface lies business.
An agent that shops for you won’t see sponsored product placements, nor will it be swayed by homepage recommendations. In 2024, Amazon’s advertising revenue exceeded $56 billion, a business built on the premise of “people browsing on Amazon.” When Amazon sued Perplexity, its complaint specifically cited the challenges posed by Agent access to ad‑impression metrics.
Amazon has its own AI shopping assistant, Rufus | Image source: GeekWire
Interestingly, Amazon does not object to the Agent itself.
It has its own shopping assistant, Rufus, which had over 300 million users by 2025. In May this year, Rufus and the upgraded Alexa+ were merged into Alexa for Shopping and integrated directly into Amazon’s main search bar. The service is also expanding beyond Amazon: the Buy for Me feature enables Amazon’s AI to place orders on other brands’ official websites on behalf of users. As of March this year, the underlying Shop Direct platform had already cataloged more than 100 million products from over 400,000 merchants.
A more subtle step occurred this April, when Amazon joined the technical committee of the Agent Transaction Protocol UCP, led by Google and Shopify.
Amazon’s logic is actually quite straightforward: it can send its agents to shop at other people’s homes, but agents from other households cannot freely shop on Amazon.
Shopify’s choice is the exact opposite, for reasons that likewise stem from its business model.
Shopify doesn’t have a super shelf that consumers visit every day; instead, it serves millions of independent stores, generating revenue from subscription fees and payment commissions rather than ad space. For Shopify, with ChatGPT, Gemini, Copilot, Muse, and any additional agent, each one simply adds another storefront for merchants—there’s nothing to protect.
According to Shopify, Shopify merchants are discoverable and purchasable by default within Muse, with product data synchronized via the Shopify Catalog. Even more crucial is the checkout process: Shop Pay already stores users’ bank cards and shipping addresses, so Muse doesn’t need to simulate login credentials for each store; instead, it uses a checkout flow that’s fully supported by Shopify.
This恰恰 responds to the core reason Amazon gave when banning Muse: compliance with security standards.
For agents operating within users’ browsers, Shopify has also prepared dedicated APIs that provide structured tools on store pages and checkout pages, enabling agents to complete orders accurately without having to parse human-readable web pages.
Amazon sells shelf space, so it aims to keep foot traffic flowing in front of its shelves; Shopify sells utilities—water, electricity, and gas—so it seeks to extend its pipelines to every corner. On the surface, the two companies’ divergence appears to be a clash between security and openness, but at its core, it reflects each business model’s most rational choice.
The emerging Agent trading network
Muse is just the latest addition to Shopify’s online ecosystem. In fact, Shopify is placing bets on three fronts simultaneously.
At the lowest level are the products. Shopify organizes over one billion items into well-structured data, complete with real-time pricing and inventory, which Agent can retrieve in just a few seconds. It has also launched the Agentic Plan, allowing brands that don’t sell on Shopify to integrate their products into the Shopify Catalog and leverage this channel to appear across various AI platforms.
The middle layer is the protocol. At the National Retail Federation’s conference in January this year, Google and Shopify unveiled the Universal Commerce Protocol (UCP), jointly developed by them, with Etsy, Wayfair, Target, Walmart, and more than 20 other retail and payment companies—including Visa, Mastercard, Stripe, American Express, and Best Buy—providing support and endorsement.
UCP Protocol Overview | Image source: Google
UCP covers the entire end-to-end process, from product discovery and adding items to the cart through checkout to after-sales service, while ensuring that merchants remain the primary parties in transactions and that customer relationships belong to the merchants. By April, Amazon, Meta, Microsoft, Salesforce, and Stripe had joined the technical committee, expanding its membership from five to ten seats.
At the top layer is checkout. Shop Pay has now processed over $400 billion in total transaction value, with 75% of users returning within 12 months, and by 2025, the average order value for Shop Pay orders is expected to be about 16% higher than that of non‑Shop Pay orders. Shopify has also launched Checkout Kit, enabling AI applications to embed the entire checkout flow—including Shop Pay—directly into conversational interfaces.
The control group for this approach is OpenAI.
In September 2025, OpenAI partnered with Stripe to launch the Agentic Commerce Protocol and introduced Instant Checkout in ChatGPT, allowing users to place orders without leaving the conversation. However, the feature was discontinued just six months later. By February this year, only about 30 Shopify merchants had actually integrated Instant Checkout, and Walmart’s data showed that the conversion rate for checking out directly within ChatGPT was roughly one-third of that for redirecting to the official website. Ultimately, OpenAI acknowledged that while users were willing to browse and compare products in ChatGPT, they were far less inclined to make payments there.
Building an in-house checkout system doesn’t work for AI companies; Shopify’s solution is to integrate merchants’ existing checkout flows into Agent. Users trust Shop Pay, which already has their bank cards saved, while merchants rely on familiar checkout rules—Agent simply handles guiding users through that final step.
If this approach proves viable, the e-commerce industry will undergo at least three changes in the near future.
The competitive landscape has shifted from “grabbing traffic and shelf space” to “whether the Agent can understand you.” Whether product data is structured, inventory is updated in real time, and rules are clear directly determines whether a product will have a chance to appear in the Agent’s recommendations. The conversion rate of catalog search is twice that of standard AI search, which already underscores this point.
Wallets and checkout will become the new moats. Whoever controls users’ identities, payment methods, and shipping addresses will be the most trusted point of handoff in agent‑based transactions. Muse has simultaneously integrated with Stripe’s Link and Shopify’s Shop Pay, and the battle at this layer has only just begun.
The e-commerce landscape will diverge into two paths: open ecosystems and walled gardens. Market forecasts cited by the organization behind UCP suggest that by 2030, agent‑driven shopping could account for 10% to 20% of U.S. e‑commerce spending, totaling between $190 billion and $385 billion. While this figure comes from stakeholders and should be viewed with some caution, the overall trend is clear.
The problem is that openness does not come without a cost.
Image source: Ignitiv
For merchants, when an Agent makes choices on behalf of consumers, the impact of carefully designed store pages and years of accumulated brand equity is diminished. Merchants may even be reduced from “brick-and-mortar stores” to nothing more than a line item of supply data in the Agent’s eyes. Ad budgets that once flowed to search engines and e-commerce platforms could also shift toward the Agent’s recommendation results. xAI previously announced that it would allow advertisers to pay for inclusion in Grok’s recommendations.
For Shopify, UCP is an open-source protocol; Stripe sits on both sides of the fence, and even Amazon has joined the technical committee. The protocol itself cannot lock anyone in—Shopify’s real leverage lies solely in Catalog data and Shop Pay users.
Of course, the era of Agent‑driven trading doesn’t bring only benefits. For the industry as a whole, the toughest challenge is accountability: if an agent places an incorrect order, makes the wrong payment, or leaks sensitive information, who should bear responsibility—the agent’s company, the wallet provider, the merchant, or the user? Musk’s approach is to make a public commitment: SpaceX will cover any errors made by the Grok Bot. But this is clearly not a solution that can be replicated across the industry.
Back to the iPhone Duo’s initial launch. In that forum post complaining about AI-driven scalping, someone raised a question: With Agents gaining traction this year, has it become even harder for real people to snag a new device?
This may be the most realistic preview of the Agent shopping era.
The agent is already the most diligent and tireless customer; every e-commerce platform must decide whether to treat it as a crawler to be guarded against or as a buyer worth courting.
Amazon and Shopify have offered diametrically opposed answers, and the next round of ranking in e-commerce may well begin with this very choice.
*Featured image source: Meta
This article is an original piece by GeekPark. For reprints, please contact Geekjun via WeChat at geekparkGO.