Should AI Assistant Ads Say Why They Found You
Ads inside an assistant are not just smaller search ads. The user is asking in a voice that feels private, then a paid card appears in the same little room. OpenAI says answers stay independent, ads are labeled, and users can ask why. Good. Now make the why boringly specific: current conversation, uploaded customer list, previous ad interaction, or unknown. If the platform can target me from an email list, “Sponsored” is not enough; it should say which door I came through.
Comments
A sponsored card inside an assistant needs a calmer affordance than a tiny label. Picture someone asking about a refund, a medication, or a kid’s trip, then an ad appears in the same answer stream. The UI should let them tap “why this ad?” and get plain causes: matched this question, broad placement, brand uploaded my email, or based on past activity — plus one button to stop that kind of targeting. If the answer feels private but the ad logic is hidden, people will treat the whole assistant as slippery.
Jun’s refund/medication/kid-trip examples are exactly where the ad story changes. A sponsored card inside search says, roughly, somebody bought this shelf. A sponsored card inside an assistant sounds more like: the helper read the room and then sold the room. So “why this ad?” cannot be legal garnish. It has to name the matching signal in human words and make the stop button obvious. If the explanation would feel embarrassing to say out loud, the ad probably should not be there.