The agentic shopping wars just split into two camps — and the fault line is exactly where you’d expect it: Amazon on one side, and almost everyone else on the other.
Amazon has blocked Meta’s new Muse personal AI agent from shopping on Amazon.com on behalf of customers. The company says it asked Meta to voluntarily exclude its e-commerce site from the agent’s shopping experience, and when that didn’t happen, it cut the agent off entirely. Users attempting to shop on Amazon through Muse since Sunday night have been met with a popup reading: “Continued access by an unauthorized AI agent violates Amazon’s Conditions of Use, to which our customers have agreed.”
Amazon’s stated grievances are specific. The company claims Meta never informed it that Muse would access its store, that the agent doesn’t identify itself while browsing, and that it appears to capture and store customer credentials — which Amazon says raises privacy and security concerns. Meta disputes this characterization, maintaining that Muse “has no visibility into people’s passwords or payment methods” and that shared credentials sit in secure storage that the agent can’t read.
The block is the latest escalation in Amazon’s campaign to keep outside agents off its site. The company has also sued Perplexity over its Comet browser and moved to block shopping agents from Google and OpenAI — a fight made stranger by the fact that Meta is simultaneously paying Amazon billions to run agentic AI workloads on its cloud.
Musk weighs in — and questions Amazon’s logic
Elon Musk was quick to poke holes in Amazon’s enforcement strategy. Reacting on X, he pointed out that if Muse accesses Amazon through the user’s own IP address and cookies, Amazon simply won’t be able to tell whether the person behind the keyboard is a human or an AI acting on their behalf.
It’s an astute observation — agentic browsing through a user’s own session is nearly indistinguishable from the user browsing themselves. Musk, who has been vocal about how quickly AI agents are moving toward doing everything a human can do with a computer, has consistently argued that the technical reality of agents will outpace any company’s ability to wall them out.
Shopify swings the other way
While Amazon builds walls, Shopify is holding the door open — and rolling out a red carpet. Shopify CEO Tobi Lutke announced that the company is “partnering deeply with Muse to enable agentic checkout with Shop Pay on all Shopify stores,” promising users “an easy and delightful way to shop and check out with Muse.”
The contrast couldn’t be sharper. Amazon sees an unauthorized intruder handling customer accounts and processing transactions without its consent; Shopify sees the future of commerce arriving early and wants every store on its platform ready for it. Shopify has been building toward this for a while — the company was already part of the launch partners for Google’s Universal Commerce Protocol, which was designed to let AI agents complete purchases across retailers. Integrating Meta’s Muse into Shop Pay gives it a second major lane into agentic checkout, this one plugged directly into the world’s biggest social graph.
Lutke’s bet also tracks with how aggressively he’s positioned Shopify around AI — the CEO has been publicly obsessed with tracking how he himself uses software, precisely so he can understand where AI agents could do the work instead.
Why the stakes are so high
The fight isn’t really about security. It’s about who owns the customer relationship — and the data that comes with it. Amazon generated more than $68 billion in ad revenue last year, a business that depends entirely on people browsing its pages and seeing sponsored products. If an agent does the browsing, filters the options, and skips the sponsored placements, Amazon’s entire advertising machine loses its fuel. Meta, meanwhile, has the opposite incentive: Muse draws on years of social data to understand taste, and the company is shipping AI capability at a furious pace to back it.
There’s also a genuine irony in the two companies’ positions. Amazon’s own agentic shopping feature, Buy for Me, browses external brands’ websites on behalf of users — but Amazon points out that it identifies itself and lets brands opt out, which is exactly the openness it’s demanding from Meta. For its part, Shopify is betting that cooperation is more profitable than control: if agents become the primary shoppers, the platforms that make it easiest for agents to check out will win the merchants.
What happens next
Amazon says it’s in direct conversation with Meta, and it has one legal lever left after losing its anti-hacking claims against Perplexity — contract and terms-of-service claims, which is exactly what the popup cites. But Musk’s point lingers: agents acting through a user’s own browser session are technically very difficult to distinguish from the user themselves, and every escalation risks a cat-and-mouse game Amazon can only win by degrading the experience for actual humans.
In the meantime, the market is voting. Muse became the No. 1 free app in Apple’s US App Store within a week of launch, ahead of ChatGPT. And with Shopify opening all of its stores to agentic checkout while Amazon locks its doors, the battle lines of agentic commerce are now drawn — not between AI companies and retailers, but between the platforms that believe agents are customers, and the ones that believe they’re trespassers.