Google's Search Dominance Faces a New Threat: Its Own AI

Google’s own AI is now costing websites 58% of their clicks from top search positions. New Ahrefs research shows the click economy that powered Google’s dominance for three decades is quietly breaking down — and it’s happening from inside the product. Here’s what’s driving it and what it means for publishers, brands, and anyone who depends on search traffic.

For nearly three decades, Google’s search dominance rested on a simple mechanic: users searched, Google served links, users clicked. That click was everything. It built an industry, funded the open web, and made Google the most powerful traffic source in internet history.

AI Overviews are quietly dismantling that mechanic.

(Chart showing the impact of AI Overviews on Position #1 CTR. Source: BusinessWire)

Ahrefs published the research on May 19, 2026, after analyzing 300,000 keywords across Google Search Console data — comparing December 2023, just before AI Overviews went mainstream, against December 2025. The headline number: top-ranking pages now see 58% lower average clickthrough rates when an AI Overview appears. Eight months earlier that figure was 34.5%. It’s correlational, not causal — Ahrefs can’t prove AI Overviews are the sole driver. But a near-doubling in eight months is hard to wave away.

Search traffic was never just a product feature. Publishers built editorial teams around it. Brands structured acquisition funnels through it. News organizations built distribution models that depended on Google sending readers their way.

That system worked because Google needed to send users somewhere. Its authority came from pointing people to the best answer, not providing the answer itself.

AI Overviews change that contract. When Google surfaces a direct answer at the top of the page, many users read it and stop. They never reach position one. The click gap Ahrefs found is not a rounding error. It’s a structural shift in how search traffic flows — and who receives it.

Here’s the tension Google hasn’t resolved. Its dominance was built on the open web — the billions of pages it indexes, ranks, and historically sent traffic to. Publishers trusted Google because Google needed them. That relationship held for decades.

AI Overviews break that balance. Google now benefits from users staying on-page. The more useful its AI answer, the less reason anyone has to click through. Publishers lose pageviews. Brands lose organic acquisition. News sites lose the referral traffic that once justified their SEO investment. And Google keeps the user and the ad impression for itself.

Google has pushed back on this framing. The company has argued that links appearing inside AI Overviews generate higher-quality clicks than standard organic links — users who do click are more intentional, and therefore more valuable. Some internal data has been cited to support this. But that argument addresses click quality, not click volume. For publishers running on pageview-based ad models, the distinction doesn’t pay the bills.

Google didn’t create this shift alone. User expectations changed first.

ChatGPT made direct answers feel normal. Perplexity made source-free responses feel acceptable. By the time Google rolled out AI Overviews broadly, millions of users had already been trained to expect a response rather than a list of links. Google accelerated that behavior. It didn’t invent it.

The practical result: users increasingly treat search as a question you ask, not a door you walk through. When Google answers well enough on-page, the door stays shut.

At Google I/O 2026, the company pushed further into agentic AI search — experiences where Google completes tasks, pulls together information, and walks users through decisions without them ever leaving the results page. Google isn’t doing this reluctantly. The product direction has been consistent for over a year now: make search self-contained, reduce the need to click out, and keep users inside Google’s environment longer. What’s less clear is how advertisers and publishers are supposed to build sustainable businesses inside a search experience that increasingly doesn’t need them.

The click loss Ahrefs found extends beyond position one. Lower-ranking pages saw meaningful drops too. That changes what good SEO measurement looks like in practice.

Rankings still matter. But a page holding position one while losing the majority of its historical clicks isn’t winning — it’s just visible. Teams should track impressions versus clicks split by AI Overview presence. They should monitor which queries regularly trigger AI answers. And they should identify where content is being cited inside AI results rather than simply ranked below them.

That last point is the practical shift. SEO has always been about position. It’s increasingly about whether the AI picks you at all.

Google processes billions of queries daily and remains the default entry point to the web for most users. That’s not changing soon.

What’s changing is the commercial logic underneath it. Ahrefs expects click loss to deepen as on-page answers become routine rather than novel. For publishers and brands, the adaptation isn’t optional — it’s just a matter of how early it starts.