Amazon Turns the Search Bar into a Canvas: Eight Visual Search Features Change How Shoppers Find Products

From AI images that materialise as you type to a lock-screen camera shortcut, the retailer is betting that pictures, not just keywords, are the future of online discovery

Amazon has detailed an expanded suite of eight visual search tools across its shopping app, signalling a broader pivot away from text-only product hunting toward an interface where images, camera input, and on-the-fly AI generation do much of the work. The features range from incremental refinements of existing tools to a headline-grabbing capability that draws AI pictures inside the search bar as customers describe what they want.

Taken together, the rollout reflects a recognition that a sizeable share of shopping intent is visual rather than verbal, and that customers often don’t know the precise term for what they’re picturing in their heads.

The marquee feature: AI images that take shape as you type

The most novel of the eight additions sits inside the search bar of the Amazon Shopping app. As a customer types descriptive language, a colour, a texture, a pattern, a silhouette, AI-generated images appear in the suggestion area beneath the search field and continuously refine with each new word.

Amazon’s pitch is straightforward: a shopper who wants a shirt with a draped collar but can’t recall the word “cowl neck,” or a sofa with woven side panels but doesn’t know to type “rattan,” can describe the look in plain language and tap the generated image that most closely matches their mental picture. The tap then surfaces visually similar products available to buy.

The feature is live for apparel and home goods, with Amazon flagging additional categories over time. It is, in effect, an attempt to close the gap between imagination and search query, a gap that has historically pushed shoppers to image-heavy platforms like Pinterest or Instagram for inspiration before they came back to Amazon to transact.

“Shop by style” collages enter search results

Sitting alongside the search-bar feature is a related apparel-focused addition. A query such as “women’s silk shirt” now returns AI-generated shoppable collages labelled “Shop by style,” grouped under themes like “Urban luxe” or “Soft elegance.”

Tapping a collage opens a curated page where shoppers can buy the pieces, explore similar items, or swipe through alternative looks. The mechanic borrows openly from social-commerce aesthetics, recasting the search results page as a magazine spread rather than a grid.

Lens Live: the camera as storefront

Amazon Lens, the company’s image-based search tool, gets its most substantial upgrade in the form of Lens Live, which turns the camera into a continuous scanning surface. As a shopper points their phone at a product, top matches populate a swipeable carousel at the bottom of the screen, with options to compare, add to cart, or save to a wish list without leaving the camera view.

Amazon has folded Alexa for Shopping, its AI shopping assistant, into the same experience. Quick product summaries appear alongside suggested questions, and a “Ask about this” tap routes practical queries to Alexa.

When the camera cannot identify a product, the tool falls back on captions describing what is visible in the scene, giving the shopper a starting point rather than a dead end.

Five more refinements

The remaining features are smaller in scope but suggest where Amazon’s product team is concentrating attention:

• Visual Suggestions. Broad queries like “flannel shirt” trigger thumbnail filters that appear while the shopper is still typing, narrowing the result set before the search has been formally submitted.
• Text-plus-image search. Customers uploading an image to Amazon Lens can now layer a written modifier on top, “like this, but in white,” or a specific brand, material, or dimension, letting the visual carry the shape while the text adjusts the attributes.
• Lock-screen widget. iOS users can pin Amazon Lens to their lock screen, opening the camera search without unlocking the phone or hunting through the app. Amazon suggests pairing it with the existing Search and Orders widgets to turn the lock screen into what it calls a “shopping hub.”
• “More like this.” A tap on any product image in search results surfaces visually similar items, for the shopper who likes a dress on offer but wants it longer, shorter, or with sleeves.
• Circle to search. Within Amazon Lens, users can draw a bounding box around a specific item inside an uploaded photo, isolating one element from a busier scene and searching for that alone.

Why all of this, and why now

Visual search is not new to Amazon. Lens has existed in some form for several years, and reverse image search is broadly available across the major retail and search platforms. What is new is the density of the rollout and the deployment of generative AI directly inside the most heavily used surface in the app: the search bar itself.

The competitive context matters. Google has been pushing its own multimodal search experiences, including a “circle to search” gesture that lent its name to Amazon’s similarly titled feature. Pinterest and TikTok Shop have built distinctive social-commerce experiences on the premise that discovery is visual first. Amazon’s response is to bring those mechanics inside the transactional environment where the purchase actually happens.

There is also a practical accessibility dividend. A shopper who lacks the specialist vocabulary of fashion, furniture, or home décor has historically been disadvantaged on a text-driven platform. Generating images from imprecise descriptions, or letting a camera do the typing, lowers that linguistic barrier.

For now, the practical takeaway for shoppers is more modest: searching on Amazon is becoming less about finding the right word and more about gesturing toward what you want and letting the app fill in the rest.

Source: https://www.aboutamazon.com/news/retail/visual-search-shopping-features