Just 13 days after launch, Muse, Meta's AI assistant, hit No. 1 on the US App Store free chart, ahead of ChatGPT, Gemini and Claude. By day 40, it had reached 10.2 million downloads in the US. JPMorgan analysts said Muse has the potential to become the most widely used consumer AI app since ChatGPT, and Meta's stock jumped more than 11% in a single day on September 21.
But the buzz had barely lasted two weeks before one giant closed its doors. Amazon blocked Muse, while another major ecommerce platform opened its checkout to it. What does that mean for online store sellers, and how should you prepare? We'll look at it in three parts.
To make sense of the split, it helps to know what Muse actually is.Meta positions Muse as a personal AI agent built for everyone. Unlike a chatbot, it doesn't just answer questions. It does things for the user, like opening a browser, sending emails, filling in forms and booking flights. That includes shopping and placing orders. Muse runs on a cloud-based virtual machine with its own browser, so tasks can keep running even after the user closes the app. Add Meta's huge distribution, and it's easy to see why it spread so fast.
Muse also draws on Meta's social network. Years of interests, activity and connections built up on Facebook, Instagram and WhatsApp act as its "social radar." Compared with a typical AI assistant, it can get a sharper read on what you want to buy, and help you buy it faster.
The same agent got opposite responses from two companies.
On September 20, Amazon began blocking Muse. When users try to buy Amazon products through Muse, a pop-up says the access comes from an unauthorized AI agent and violates Amazon's conditions of use.
Amazon gave three reasons. Muse wasn't authorized, it didn't identify itself as an AI, and Amazon was concerned about customer credentials. Meta's response is that Muse can't see users' passwords or payment methods.
Analysts see a deeper reason tied to advertising. Amazon's advertising revenue reached $68.6 billion in 2025, and that business depends on people browsing. They search, scroll past sponsored products and get nudged by algorithms. An agent doesn't browse. It finds what the user wants and checks out, so ads get fewer chances to show.
Another major ecommerce platform went the other way. On the day Muse launched, it added Meta as an AI sales channel. On September 21, the two companies announced that Muse would support checkout on that platform's stores through its wallet payment, so users can pay right inside the conversation.
Each choice follows the company's own business. Amazon runs its own storefront, recommendations and advertising, so an outside agent that bypasses them also bypasses its entrance. A platform built for independent stores cares more about where the transaction is completed than where the traffic comes from.
The same logic applies to online store sellers, so the next question is what you can do about it.
In the age of AI agents, entry points are being redistributed. The competition is no longer only about search rankings and ad slots. It's also about who AI can call on and recommend. Even without connecting to any protocol, the door isn't closed. According to a Forkast report, Muse has three checkout paths. They are that platform's wallet checkout, Stripe Link, and a built-in browser that operates web pages directly as a fallback. In other words, online stores are naturally within an agent's reach.
The catch is that AI has to be able to "read" you. That's the problem GEO (generative engine optimization) is meant to solve. Traditional SEO competes for rankings in search engines. GEO is about helping AI understand you and choose you when it finds products, compares options and makes decisions for users. AI often makes those decisions within seconds. It reads page information and compares specs, prices, reviews and fulfillment. Your product specs, size charts, shipping and returns policies and brand story are no longer written only for people. They are also decision material for AI.
There's another upside. This traffic isn't bought through bidding. It comes from AI recommendations. Shoppers arrive with purchase intent, so the path to conversion tends to be shorter. For online stores that keep worrying about traffic costs, this is a new entry point worth preparing for early.
To catch this wave, you can start with three things in your existing store.
Write out titles, specs, materials, use cases and size charts in full. AI can then extract everything in one pass, and your products won't be skipped because of missing information.
When you have a lot of products, filling this in by hand takes time. You can leave it to Shoplazza Athena, an AI assistant for ecommerce built into your admin. Send it product images, links or an Excel sheet, and it creates products in bulk, generating titles, descriptions, tags and SEO information. It shows a preview card first. For key details like price, material, size and stock, we suggest checking each one against your real product data before you apply. If products are already listed and need more detail or a consistent format, you can ask Athena to edit them in bulk too.
Product pages, FAQs and shipping and returns policies should be written in a way AI can read, with a clear structure. AI often weighs fulfillment too. If shipping rules and free shipping thresholds are vague, it's harder for AI to tell whether your store is a good fit.
You can tell Athena this part directly as well. It can look up your existing shipping plans and update shipping fees and free shipping thresholds in one go. You can also create delivery zones and shipping plans through conversation, and they take effect once you confirm. Shoplazza's built-in SEO features, such as structured data and automatic image ALT text, also make pages easier for machines to read.
Keep your brand story, media coverage and expert content consistent across the web. When AI looks for supporting evidence, that's how it finds your credibility. This takes time to build, but the starting point can be simple. Make your website's brand description match how you're described on outside platforms.
Once the three things are done, data insights can show you what to do next. Athena Skills turn the operating experience of hundreds of thousands of online stores into diagnostic tools you can use directly. One of them, checkout conversion diagnosis, compares your store against industry benchmarks for the same country and order value range, and finds where shoppers drop off at checkout. You need to activate Shoplazza Payments first. Orders that come from AI still have to be completed at checkout, so it's worth checking this step early. None of this means redoing your SEO from scratch. Most of the work is making the product and page information you already have more complete and clearer.
Shoplazza is part of this wave of agentic commerce. This year it launched Athena, an AI agent that breaks tasks down and coordinates store building, visual content, marketing and customer retention. Athena Skills turn the operating experience of more than 650,000 merchants into capabilities that AI can call. The store itself is becoming an "AI-ready" asset.
Amazon's block can't stop agents. It only shows one thing, that entry points are shifting. The sellers who get their stores ready for AI to call on first will be the first to reach the first customers.
They sit on opposite sides of a purchase. Muse is Meta's personal AI agent. It browses stores, compares options and places orders for shoppers, so it works for the buyer. Athena is the AI operations agent inside the Shoplazza admin. It works for the merchant, helping create products, set up shipping, look up orders and analyze data. One buys, the other sells. When merchants use Athena to organize their product information and shipping rules, their stores are easier for AI like Muse to read and pick.
When AI picks products for a shopper, it relies on the titles, specs, materials and size charts on your pages. Products with missing information tend to get skipped. Filling these in by hand is manageable with a few products, but hard to keep up when you have many. This is where AI bulk product listing helps. Send Athena product images, links or Excel or CSV files, and it creates products in bulk, generating titles, descriptions, tags, SEO information and price suggestions. A preview card appears before anything is created. Check price, material, weight, size and stock against your real product data one by one, then apply.
When AI compares prices and makes decisions, shipping fees and free shipping thresholds are key signals of fulfillment ability. If they're vague, it's harder for AI to tell whether your store is a good fit. To configure shipping and shipping plans with AI, just tell Athena what you want. It can look up your existing shipping plans and update shipping fees and free shipping thresholds in one go. You can also create delivery zones and shipping plans through conversation, including whether cash on delivery is supported and whether pricing is based on a fixed amount, order value, weight or quantity.
No. For important actions like creating, editing and deleting, Athena shows a preview first and only runs the action after you confirm. So before any bulk change, you can see exactly what will change, then decide whether to apply it.