OpenAI is testing colored product labels inside ChatGPT shopping results, including a green "best all-rounder" chip and gray badges for criteria such as "best for beginners" and "best cushioning," according to search analyst Brodie Clark.

Clark posted screenshots of the test on X and to his SERP Alert subscribers, framing the colored tags as a click-through lever for organic product listings inside the chatbot.

What do the labels actually show?

The new variant layers two visual tiers on top of the existing product panel. A single item in any given response carries the green "best all-rounder" badge. Other products pick up gray labels tied to narrower positioning criteria, mirroring the comparison language shoppers already see on retail sites.

Clark's description: "ChatGPT is testing out a new variation of the labels that can be displayed for organic products in its results. Expanding upon the original version, the new experiment is displaying a green label for the 'best all-rounder' product and also additional grey labels for the 'best' for different criteria."

The labels attach to the recommendation surface rather than to a single named retailer, which keeps the test framed as organic product highlighting.

How does this compare to the earlier ChatGPT label experiment?

The current build is the second visible version of the labels. Clark described the original as a plainer, single-tag format pointing to one product. The new variant adds the green chip to the headline pick and expands the gray labels across the rest of the listing.

In practice the layering turns a flat list into a comparative shortlist. Shoppers can scan criteria at a glance instead of reading each product card line by line.

Why does a label color matter for click-through?

Color draws the eye first, and a green chip on one item concentrates attention on that entry. "This should help with click-through rates on these products," Clark wrote, calling the design an explicit CTR lever.

The labels do not reference revenue mechanics in the screenshot, but a colored highlight introduces a paid-feeling cue inside a recommendation panel that started as plain text. That design choice matters for any retailer whose products already appear in ChatGPT shopping answers.

Who spotted it, and where did the report originate?

Brodie Clark runs the SERP Alert newsletter, which tracks Google and AI-search SERP changes. He posted screenshots to X and to SERP Alert subscribers. Forum reaction moved to the X thread, where several AI-search watchers compared the design to Amazon's "Choice" badge and Google's product result annotations.

Clark's coverage typically focuses on Google features, which gives his read of a ChatGPT-specific label extra signal. AI-search label tests rarely surface from Google-focused analysts unless the visual change is clear-cut.

What should retailers and SEOs monitor next?

• Whether the labels persist past the A/B test window.
• Whether the criteria expand beyond "all-rounder," "beginners," and "cushioning" into other verticals.
• The CTR gap between green-tagged, gray-tagged, and unlabeled products, which OpenAI has not disclosed.
• Any merchant-feed or schema signal that would let brands optimize for a specific label.

For now, Clark's screenshots are the only confirmed sighting on the record. Watch his SERP Alert feed and the linked X thread for replication across more shopping queries before treating the labels as a stable feature in ChatGPT.