Google removes some AI health summaries after investigation finds “dangerous” flaws

Google removed some AI health summaries due to inaccuracies that could pose risks to users, following a Guardian investigation. The AI feature provided misleading health information, potentially leading seriously ill patients to believe they are healthy. Specific queries like "what is the normal range for liver blood tests" were disabled after experts flagged them as dangerous. The investigation revealed flaws in providing essential context and adjusting figures for patient demographics, potentially causing patients to skip necessary follow-up care. Despite the removals, Google stated that the majority of its AI health summaries are accurate and supported by high-quality sources.