Google AI Search Now Hits 43% of Queries-Why 60% of Clicks Are the Real Story

AI Overview penetration is high, but the key question is where the clicks go

The headline number is already striking: AI Overviews now appear on 48% of queries, and in the U.S. that rises to about 50% of queries. But the more important question is simpler: if Google is answering so much of search directly, where do the clicks - and the monetizable attention - go?

That is the core debate. Google clearly sees momentum in its new setup: AI Mode has surpassed one billion monthly users, queries have reached an all-time high, and the company is putting AI tools directly in the Search box. At the same time, the user behavior behind AI Overviews still points to fewer clicks downstream: when an AI Overview triggers, organic CTR can fall by up to 61%.

My view is that the market should focus less on the feature headline and more on the transfer of attention. If Google keeps capturing the answer at the top of the experience, the real question is not whether AI search is real. It is who captures the value attached to that answer.

AI Overviews change the search funnel by giving Google the first answer

The pressure point is downstream traffic

The scale is already clear. In the U.S., 50% of queries trigger AI Overviews, and when that answer layer appears, organic CTR can fall by up to 61%. That is the mechanism investors and publishers need to price: less traffic flows to traditional results, while more value concentrates around whether a property appears inside Google's own summary.

Citation matters more when Google answers first

This is not just a traffic tax. It changes what visibility has to look like. Brands that get cited inside AI-generated summaries earn 35% more organic clicks than those that are not. That means ranking in the blue links still helps, but citation can be the difference between capturing traffic and losing it once the AI answer sits above the results.

The source mix also matters. Google's AI Overviews often pull from Reddit (21%) and YouTube (18.8%), which favors discussion, video explanation, and fast-moving content over static pages. In that environment, the role of AI Overviews is only going to increase, and the winners will likely be the sources Google relies on to resolve queries quickly.

AI Mode may operate differently from classic Search

The more important structural question is AI Mode. Experiments suggest it draws from a proprietary content store separate from the search index, rather than the live web in the classic Search sense. If that holds up, ranking alone is no longer the full strategy; visibility inside Google's broader answer layer matters just as much.

Regulation is the main external risk to the rollout

For investors and publishers, this is as much a policy story as a product story. Google is already responding to criticism that AI answers are reducing traffic to the open web by adding Further Exploration links, subscription labels, and inline link context. The pressure is real: AI Overviews have been linked to a 58 per cent reduction in click-through rates, so Google cannot treat this as a purely product-led rollout.

How to frame the next 12 months

The near-term setup still looks constructive for Google, but it is not risk-free. The EU complaint means regulators may impose an interim measure to prevent what they call irreparable harm while the case is investigated, which could slow rollout or create headline risk in Europe. In the U.S., the recent ruling left Google's monopoly finding in place but did not materially overturn the core search economics in a market-breaking way.

For investors, the framework is straightforward: - Bullish signal: Google keeps expanding answer-led search while new publisher-facing features help preserve the content loop. - Bullish signal: Publisher economics stabilize because Further Exploration links and related UI changes drive more traffic back to sources. - Bearish signal: Europe forces a temporary limit on AI Overviews while competition and content concerns are investigated. - Watchpoint: Markets keep treating the U.S. ruling as noise because the remedy remains limited. - Invalidation: If cited sources do not convert AI visibility into durable traffic, and richer answers continue to pull attention away from traditional links, the AI Search thesis gets harder to defend.