Do all Google searches use AI now?

If you've watched your Google traffic dip this year, you've probably asked the question with a knot in your stomach: did Google just replace regular search with AI — overnight?

The headlines certainly make it feel that way. AI Overviews now appear in over 200 countries and 40+ languages, and Google has launched a separate AI Mode where the entire results page is generated by AI. Meanwhile, NPR reports that CNN's traffic is down roughly 30% year-over-year, Business Insider and HuffPost are down around 40%, and one travel blog shut down entirely after losing 90% of its visitors. No wonder business owners assume the old search engine is dead.

But here's where the confusion sets in. There are actually two very different things happening at once, and most coverage blurs them together:

• AI you can see — AI Overviews (the summary boxes at the top of results) and AI Mode (an opt-in, fully AI-generated results page)
• AI you can't see — machine learning and ranking systems Google has quietly used behind the scenes for years
• The assumption that one means the other — that visible AI everywhere means traditional search is gone

The visible stuff is still a minority experience. According to Pew Research's analysis of nearly 69,000 real searches, only about 18% of Google searches in March 2025 produced an AI Overview. Independent Semrush Sensor data puts the figure at roughly 15% of keywords. So no — not every search is an AI search, at least not in the way the headlines imply.

The invisible stuff is a different story, and it's actually reassuring. Google's own documentation states that its generative AI features are "rooted in our core Search ranking and quality systems." Translation: AI Overviews and AI Mode are a layer built on top of the traditional index — the same index your website is ranked in today. Google even confirms that SEO best practices remain fully relevant for AI features.

That matters for how you invest in getting found. Whether you handle SEO yourself or work with a partner like AI Business Sites, the foundation hasn't changed: indexed pages, strong rankings, and well-structured content are what make you visible to both the traditional results and the AI answers layered on top.

The rest of this article unpacks exactly how much of Google is AI now, what triggers an AI answer versus a classic results page, and what the data says you should actually do about it.

The short answer is no — and the numbers prove it. In March 2025, only about 18% of Google searches produced an AI Overview, based on a Pew Research study of 68,879 real searches from 900 U.S. adults. Independent data from Semrush's Sensor puts the figure at roughly 15% of keywords. NPR's reporting on the same period lands at about 20%. Three different measurements, one consistent conclusion: the vast majority of searches still look like traditional Google.

But here's the nuance that matters: while AI Overviews appear on a minority of searches, they're hard to avoid entirely. That same Pew study found that 58% of U.S. adults encountered at least one AI summary during their searches. And the reach is expanding fast — AI Overviews are now available in more than 200 countries and 40+ languages, covering every country except Cuba, Iran, and mainland China.

So why does it feel like AI is everywhere? Because what you search determines what you see. The odds of getting an AI summary vary wildly by query type:

• 60% of question-based queries (who, what, when, why) triggered an AI summary
• 53% of searches with ten or more words produced one
• Only 8% of one- or two-word searches showed an AI Overview
• Just 15% of direct local-intent queries (like "plumber near me") got one

That last data point comes from Whitespark's analysis of 540 queries, which found a stark split: only 15% of local-intent searches showed AI Overviews, while 92% of informational queries and 97% of hybrid queries did. For a small business owner, this explains a lot of confusion. If you search "roof repair cost" you'll likely see an AI summary. If you search "roof repair" you probably won't. Your personal experience of Google depends heavily on how you type.

The takeaway for anyone relying on Google to get found: AI summaries are selective today, but they cluster exactly around the informational and question-based searches that bring customers into a buying journey — and when they appear, click-through to websites nearly halves, dropping from 15% to just 8% of visits. That's why we track both traditional rankings and AI citation visibility for every AI Business Sites client: being findable in classic results and being recommended by AI are now two different jobs, and the businesses that win treat them that way.

When Google shows you an AI summary, it isn't replacing the search engine underneath — it's reading from it. That single distinction changes everything about how your business should approach AI visibility.

Google's own Search Central documentation is unambiguous: "our generative AI features on Google Search are rooted in our core Search ranking and quality systems." The AI layer uses retrieval-augmented generation to pull pages from the existing index, running "query fan-out" to explore multiple related searches simultaneously. In other words, if your page isn't indexed and ranking well, the AI can't cite it.

That's not just Google's opinion. A Semrush study of 5,000 keywords and 150,000+ citations found that strong organic rankings correlate strongly with AI visibility across every LLM studied. Their conclusion: AI search isn't replacing traditional SEO — it's creating its own category on top of it. Traditional SEO remains the gateway to appearing in AI answers.

Google also takes direct aim at the cottage industry of "AEO hacks." The same documentation debunks llms.txt files, special AI markup, and content chunking as unnecessary — if a service promises to "optimize you for AI" separately from search fundamentals, that's a red flag. What works is the unglamorous stuff: crawlable pages, clean structure, snippets, and genuine topical authority. This is why AI Business Sites builds visibility the same way for both — solid SEO foundations that AI features inherit automatically.

The honest part of the picture is the traffic math. Pew Research, analyzing 68,879 real searches from 900 U.S. adults, found that users clicked a traditional result on just 8% of visits when an AI summary appeared — versus 15% without one. And only 1% clicked a cited source within the AI Overview itself. When AI answers the question at the top of the page, many searchers never scroll past it.

What this means practically:

• Ranking well is now doubly important — it drives clicks directly and determines whether AI cites you at all.
• Being mentioned in an AI answer has brand value even when the click doesn't follow.
• Query type matters: Whitespark's analysis of 540 queries found AI Overviews on 92% of informational queries but only 15% of direct local-intent searches.
• Diversifying beyond Google clicks — email, direct traffic, reviews — is no longer optional insurance.

The businesses that win the AI era will be the ones that never abandoned search fundamentals — they simply built deep enough authority that the AI layer had no choice but to cite them.

If a customer asks an AI assistant "who's the best HVAC company near me?" and your business isn't in the answer, that customer never sees you. That's the new reality of search — and the data shows exactly where visibility is won and lost.

The good news: AI visibility isn't random. It follows content patterns you can build for. According to Pew Research Center's analysis of nearly 69,000 real searches, 60% of question-based queries trigger an AI summary — versus just 8% of one- or two-word searches. And Whitespark's local-search study found AI Overviews appear on 92% of informational queries and 97% of hybrid queries, but only 15% of direct local-intent searches.

The takeaway for a local business is clear: question-answering content is where AI visibility is won. Pages that explain how much a furnace repair costs, how long a roof lasts, or what to do about a flooded basement are the pages AI systems pull from when they build answers — and cite.

Here's the catch: when AI answers, it narrows the field dramatically. That same Uberall benchmark study found AI assistants typically recommend only 3–5 brands per query. Being ranked on page one isn't enough anymore — being one of the handful of names the AI actually says out loud is the new prize. Miss the cut, and a competitor gets the mention instead.

Then there's the measurement gap. Google's own AI reporting tools are limited — its Generative AI performance report in Search Console lacks a query dimension and mixes AI data into overall totals, so you can't easily see which searches surfaced you in AI answers. And since click-through nearly halves when an AI summary appears (8% versus 15%), rankings alone tell you less and less. You need to know whether AI assistants recommend you — across ChatGPT, Perplexity, and Google's AI Mode — not just where your pages rank.

For a busy owner, that translates into three ongoing jobs:

• Publish consistent, question-based content that matches how customers actually ask
• Keep traditional SEO strong — Google confirms its AI features are rooted in core ranking systems, so rankings gate AI citations
• Track AI-citation share of voice against competitors, since Google's tools won't show it clearly

This is exactly the work a managed site handles on autopilot. AI Business Sites builds question-answering service and location pages at launch and publishes fresh keyword-targeted content every month, while a built-in SEO team monitors whether AI assistants across ChatGPT, Perplexity, and Google's AI Mode are recommending you — with a weekly report landing in your inbox. The owner runs the business; the system runs the visibility.

So no, not every Google search runs on AI — but the direction is clear, and the smart move is preparation, not panic. With only about 18% of searches producing an AI Overview as of March 2025, traditional search still carries most of the traffic. That window is your runway.

The single most important thing to understand: traditional SEO is not being replaced by AI visibility — it gates it. Google's own AI optimization guidance confirms its generative features are "rooted in our core Search ranking and quality systems," and Semrush's research found strong organic rankings correlate strongly with AI citations. Skip the "AEO hacks" — Google explicitly says llms.txt files and special AI markup aren't used.

Here's your practical playbook:

• Keep investing in SEO fundamentals. Indexed pages with clean structure and snippets are the entry requirement for AI features — rankings and AI visibility rise together.
• Publish content that answers real customer questions. Pew found 60% of question-based queries trigger AI summaries, versus just 8% of one- or two-word searches — question-answering content gets surfaced and cited.
• Diversify beyond Google clicks. Click-through nearly halves when an AI summary appears (8% vs. 15%), and publishers are already seeing 30–40% traffic declines. Email lists, direct traffic, and owned channels are your hedge.
• Track your AI citation share-of-voice, not just rankings. AI assistants typically recommend only three to five brands per query — you need to know if you're one of them.

That last point is where most small businesses get stuck. Google's own reporting on AI visibility is thin, and watching two fronts at once — traditional rankings and AI citations — is a real operational burden on top of running the business.

This is exactly what AI Business Sites handles. Every site we build and manage runs a weekly SEO loop that tracks Google and Bing rankings alongside AI-citation share-of-voice across ChatGPT, Perplexity, Google AI Mode, and Gemini — then publishes 14+ keyword-targeted content pieces a month that answer the questions your customers actually ask. One system, both fronts, done for you.

Want to see what that looks like for your business? Book a free strategy call — 30 minutes, no obligation, no sales pitch. We'll look at where you stand today and map out how a done-for-you site and SEO engine covers both traditional rankings and AI visibility.