An investigation into the fate of Google’s “What People Suggest” feature reveals a story of a health AI product launched with ambition and retired with minimal accountability. The tool used AI to curate community health perspectives from online discussions and display them in search results. Three insiders confirmed it is no longer operational, and Google acknowledged its removal while providing an unconvincing explanation.
Launched at Google’s “The Check Up” health conference in New York, the feature was introduced by then-chief health officer Karen DeSalvo as a way to enrich health search with authentic peer experiences. She wrote that users often search for health information not just from experts but from people living with the same conditions. The AI-organized community content was initially available to mobile users in the United States.
Google cited search page simplification as the reason for the removal and explicitly stated that safety was not a factor. When pressed for public documentation of the decision, the company pointed to a blog post that made no mention of the discontinued feature. This transparency gap has been widely flagged.
The episode connects to broader scrutiny of Google’s AI health tools. An investigation earlier this year found that AI Overviews were providing false medical information to roughly two billion monthly users. While Google responded by removing AI Overviews from some health searches, health advocates described the measure as insufficient and called for deeper reform.
At its next health event, Google will attempt to frame its role in health AI positively. Whether it can do so credibly will depend on whether the company demonstrates a genuine commitment to accountability — including openly acknowledging and learning from experiences like the one surrounding “What People Suggest.”
Inside the Removal of Google’s AI Health Feature That Relied on Public Opinion
