Google AI Overviews Wrong 1 in 10 Cases, New Analysis Finds
Roughly 10% of Google’s AI Overviews—the summaries stacked above standard search listings—got it wrong, according to a fresh New York Times-commissioned analysis, despite Google shifting the feature to its newer Gemini 3 model. AI startup Oumi ran 4,326 queries and saw accuracy climb from 85% in October to 91% by February. Now it's significant: Google is weaving AI-generated answers right into core search, not just leaving them as a side feature. Back in January, Gemini 3 became the default for AI Overviews, and the company introduced a follow-up question function. By late 2025, executives said more than 2 billion users were on board.