Google's AI answer boxes seem to make people click through to websites way less often — but is AI really the cause?

This explores whether AI answer summaries at the top of search results are the reason more searches now end without a click, and how strong that link is. For any single search, the corpus says yes, clearly. Pew's browsing-panel data on nearly 69,000 real Google searches found people clicked a result link 8% of the time when an AI summary appeared, compared with 15% when none did. Searches with a summary were also 10 percentage points more likely to end with no click at all Do AI summaries on Google reduce clicks to actual websites?. Ahrefs found a similar pattern from the publisher's side: for keywords that trigger an AI Overview, the top-ranked result got about 58% fewer clicks How much are AI Overviews actually reducing organic search clicks?. That Ahrefs figure is a correlation, not a controlled test, but it points the same way as Pew.

The mechanism is simpler than you might expect. An eye-tracking study from RMIT and Microsoft found the AI Overview now takes the 'golden' spot on the page where people's eyes go first. The share of attention going to the first-ranked link fell from 31% to 9%. People also rated the summary and the ranked links as equally trustworthy Where do searchers look when AI Overviews appear?. So readers aren't avoiding websites out of distrust. The summary takes their attention and answers the question before they reach the links. An experiment outside search shows this is close to built in. Nextdoor tested AI-written notification summaries that were measurably more informative, and they *lowered* click-through, because the summary already gave people what they needed Does better summary writing actually increase user engagement?. A better summary leaves less reason to click, so the drop isn't a flaw that can be engineered away.

The size of the effect depends heavily on where you measure it. A natural experiment comparing English Wikipedia with its German and French editions put the traffic loss from AI Overviews at only about 5% Does AI search summaries divert traffic away from Wikipedia?. That is far smaller than the 58% drop at the top commercial result. News publishers are bracing for something much bigger: in a Reuters Institute survey, executives expected search referrals to fall 40–43% over three years. The same report says the measured decline has started but hasn't yet been sized Will AI Overviews reduce search referral traffic to publishers?. One caveat about the corpus itself: it shows strongly that a summary reduces clicks on the searches where it appears. It doesn't contain a long-run count of zero-click searches across the whole web, so it can't say how much of any overall rise comes from AI rather than older features like featured snippets and instant answers.

Habits are also shifting more slowly than the traffic numbers suggest. Nielsen Norman Group found that people using AI chat kept using traditional search throughout their tasks, often running both side by side. AI is being added to search, not replacing it Does generative AI chat actually replace traditional search?.

The part you might not have thought to ask about is what readers lose when they don't click. Seven randomized experiments with more than 10,000 people found that those who learned a topic from an AI summary instead of web search reported learning less and felt less ownership of what they knew. The advice they later wrote was rated sparser and less useful Does learning from AI summaries produce shallower knowledge than web search?. So zero-click searches cost more than publisher traffic. Clicking through to sources and comparing them is part of how people come to understand a topic, and a summary skips that step.