From Launch to Shutdown: Google’s AI Amateur Medical Advice Tool Is No More

What began with a confident launch and ambitious promises ended with a quiet exit. Google’s “What People Suggest” feature, which used AI to compile health advice from online strangers, has been discontinued. The fact of its removal was confirmed by three insiders and subsequently by a Google spokesperson, though the company’s account of why it was removed remains incomplete.
The feature was introduced at “The Check Up,” Google’s annual health event held in New York. Karen DeSalvo, then the company’s chief health officer, framed it as a meaningful innovation that would help users discover health wisdom from people who had lived through similar experiences. The AI curated community forum content into navigable themes for users with health questions.
Google said the removal was a routine simplification of the search page and explicitly denied it had anything to do with quality or safety. Yet the evidence offered for a public announcement turned out to be a blog post that never referenced the feature. Critics pointed to this contradiction as emblematic of how Google manages uncomfortable truths.
The story unfolds against a year of mounting concern about AI health content. An investigation earlier this year revealed that Google’s AI Overviews were sharing inaccurate health information with billions of users monthly. The removal of some medical AI Overviews following that investigation did not fully restore confidence in Google’s ability to responsibly handle health information.
As Google moves forward with its health AI ambitions, it must grapple with the legacy of “What People Suggest.” Products that touch on health and medicine demand extraordinary care, and the lesson here is clear: if a tool is removed due to problems, the public deserves to know. Accountability is not optional when health outcomes are at stake.