Amazon introduced its “About You” feature in May as part of a broader AI-shopping push. The page compiles inferences from purchases, searches, saved lists, reviews and conversations with Alexa for Shopping, then lets customers edit or remove details that influence recommendations.¹

The idea is straightforward: make Amazon’s immense catalogue easier to navigate by telling the retailer what it has guessed about you. But the feature drew renewed attention this week as screenshots spread across Threads and Reddit, where users discovered that the algorithm can sound less like a helpful assistant than an overly familiar relative.

One shopper said Amazon had concluded she had “flat buttocks”; another found an entry stating that she had a husband whose “skinny legs” had apparently also become part of the retail record. A third was tagged as someone who uses a phone in the shower to read Kindle books — an inference the user conceded was accurate, if unnerving.¹

The viral jokes mask a more complicated result. Amazon’s profiles often appear capable of spotting recurring purchases, pets, neighbourhood habits and food preferences. In an informal Business Insider newsroom check, accounts typically produced 30 to 45 observations, with many transaction-based conclusions landing close to the mark.²

Yet the leap from shopping data to a coherent human portrait remains shaky. Shared accounts muddled household details; several users said Amazon got the number of children wrong, while one editor’s profile invented a girlfriend named “Barry.” Another was labeled a beef-jerky enthusiast despite buying far more energy bars and vitamins, calling the result “a judgy accusation from a nutritionist.”²

That tension is what has made About You travel so quickly online: the tool offers users control over its entries, but also exposes just how much interpretation sits behind a supposedly simple product recommendation.