Earlier this summer, Amazon customers began noticing that emails related to their online orders looked sparse: Order confirmation emails didn’t name specific items anymore, and instead listed only item categories.
“Your Beauty item is confirmed!” an email about my retainer cleaning tablets read. Shoppers have posted other iterations of the redacted emails as well: “Ordered: 1 Hardware item,” “Your Drugstore, Shoes, and other items are here!” and “1 Nutrition & Wellness, 1 Wireless Accessories,” for example. The emails have clip art-style illustrations of general product categories, and a shopper has to exit their email and go to Amazon to see what item it’s actually referring to.
Complaints about the vague emails appear to have started in July; orders placed as recently as June contain thumbnails and names of the exact item purchased. Some shoppers are complaining to Amazon about the new system, saying the emails with no details make it harder to keep track of purchases (and kind of look like spam or a phishing attempt). Amazon has a long history of tweaking how order emails look — there are complaints going back years about what’s included and what isn’t — but in the last month and a half, more and more shoppers are noticing the vague order update notifications.
AI is likely partially to blame
Over the last year or so, tech companies have asked consumers to imagine a world where an AI “agent” is doing their bidding for them, including spending their money. Google, with a catalog of billions of products in Shopping, is perhaps best situated to make that vision a reality, and indeed has already introduced a slew of shopping features powered by AI. Gemini Spark can mine a user’s email for crumbs of information to, for example, build custom vacation itineraries, pulling in data like concert ticket confirmations and biographical details about family. In May Google unveiled a shopping cart that works across various merchants that would eventually be integrated into Gmail as well, meaning shoppers could add things to their cart while reading their inbox or watching YouTube. How helpful an AI agent can be depends on how much personal information a user is willing to fork over.
As users hand over more and more data and agency to AI tools, service providers are caught in a bind: If agents are truly as efficient as tech companies say they are, customers and audiences in theory have no reason to visit a website themselves — the agent can just do it for them. This phenomenon has been called “the DoorDash problem.” Still, many big retailers, like Walmart, Target, and Wayfair, have signed on to commerce features built into Google’s AI surfaces, hoping that consumers will research, compare, and ultimately purchase their products via chatbots. Notably not on the list of merchants: Amazon.
Amazon’s protective stance
Amazon has been more protective of its direct relationship with shoppers. Last year the company sued Perplexity for allowing users to shop on Amazon on their behalf; a judge sided with Perplexity earlier this month. Amazon has also been plugging away at AI-powered shopping features of its own, like price tracking and automatic purchasing. Alexa for Shopping is built into the Amazon app, and shoppers can query the chatbot for product recommendations and comparisons without going outside to Gemini or ChatGPT, for example.
When asked directly if the change in email format was to shield purchase information from companies like Google that are building AI-powered shopping tools, Amazon spokesperson Maxine Tagay said in an email: “As customers shop with us more frequently, including on their phones, and use the ‘Your Orders’ page in the Amazon app to get real-time, consolidated order details and delivery status, we’ve simplified several order-related emails to direct customers to our app and website for the latest information on their orders. This also reduces customer information shared outside the Amazon app and website to further improve customer privacy.”
In other words: It’s a way to hedge against the DoorDash problem by hiding the meaty information on Amazon’s properties, and to keep customers visiting the retailer directly. The upshot for shoppers is that it’s ever so slightly more annoying to access our own information, too.
A history of email tinkering
Amazon’s order emails have never been static. Over the years, the company has repeatedly changed the design, content, and frequency of its transactional messages. Some changes were subtle, like moving the order total or delivery estimate. Others were more drastic, like removing item images or consolidating multiple shipments into a single email. Each time, a subset of customers has voiced confusion or frustration, only for the change to become the new normal. The latest shift is notable because it strips out the very information that makes an order confirmation useful: the identity of the product itself.
For many shoppers, the email serves as a lightweight receipt and a way to track what’s coming. Without specific product names, the email becomes little more than a nudge to check the Amazon app. That’s by design, according to the company. Amazon wants to be the destination for all order-related activity, not just the place where you click “buy.” This is part of a broader strategy to make the Amazon app indispensable for the entire post-purchase journey.
Privacy and the AI agent race
Amazon frames the change as a privacy win. By keeping detailed order information within its own ecosystem, the company argues, it reduces the risk of sensitive purchase data being scraped by third parties. That rationale has some merit. Email is a notoriously insecure medium, and third-party services many people use to parse inboxes — such as search engines, AI assistants, or even email management tools — can access order details. For high-ticket items or sensitive health-related purchases, the category-only approach does add a layer of obscurity.
But critics note that Amazon is also protecting its commercial turf. AI agents from Google, OpenAI, and others are increasingly capable of placing orders on a user’s behalf. If those agents can read confirmation emails and extract not just the item name but also the price, seller, and delivery date, they can offer users a seamless shopping experience without ever visiting Amazon. The retail giant has spent decades building one of the most efficient direct-to-consumer platforms in history. Surrendering control of the post-purchase experience to outside AI tools would erode a key competitive advantage.
Amazon has already demonstrated its willingness to litigate against AI companies that enable on-platform shopping through agents. The Perplexity lawsuit, though recently ruled in the AI company’s favor, signals Amazon’s intent to act aggressively to preserve its relationship with shoppers. The email change can be seen as another tool in that same campaign — one that doesn’t require legal action but instead alters the data environment on which AI agents rely.
What it means for shoppers
For ordinary customers, the practical impact is mild but real. If you rely on email to keep a paper trail of purchases, you now have to log into Amazon’s app or website to see exactly what you bought. That extra step is a nuisance, especially for people who manage many orders each month. The new format also makes it harder to spot fraudulent activity at a glance. A vague “Hardware item” notification could be a legitimate purchase or a sign that an account has been compromised. This has led some security researchers to worry that the change could reduce the effectiveness of email-based transaction monitoring.
There are also accessibility considerations. Some users prefer email over apps because of visual impairments, low bandwidth, or simply a desire to avoid clutter. For those users, the category-only emails provide less value. And because the emails look more generic, they also resemble spam — a dangerous quality in an era of phishing attacks. Amazon’s own anti-phishing guidance has long told customers to check for specific order details to identify legitimate emails. Now those details are missing, making it harder for customers to distinguish real notifications from fakes.
Despite the backlash, the change is unlikely to be reversed. Amazon has a track record of rolling out unpopular features and ignoring complaints if those features support a larger strategic goal. The company is betting that as AI agents become more common, the companies that control the original transaction flow will retain the most value. By withholding granular order data from email and forcing users into its app, Amazon is making itself the only reliable source for order information — and, by extension, the only merchant that can safely participate in the AI-powered shopping ecosystem of the future.
For now, shoppers can still find their complete order details in the “Your Orders” page or the Amazon app. The information hasn’t been deleted; it’s just less accessible. But every extra click, every login prompt, and every vague email nudges customers ever closer to Amazon’s own walled garden. That may be good for Amazon’s bottom line. Whether it’s good for consumers is another question.
Source:The Verge News