How AI Is Changing SEO: What Businesses Need to Know
A lot of what’s written about AI and SEO is forecast presented as fact. We kept the two apart, using Pew Research, Google’s own statements and our tracking data from 103,427 AI answers.
You’ll find the five changes that matter most and what stays the same. After that come the organic traffic debate and a seven-step plan for what to do now.
Key takeaways
- In a Pew Research Center study, only 1% of visits with an AI summary included a click on a link inside the summary.
- Alphabet said in July 2026 that AI Mode passed 1 billion monthly active users and that AI features send billions of clicks to websites every week.
- Google says AI Overviews and AI Mode are rooted in its core Search ranking systems, so SEO fundamentals still decide which pages can be cited.
- ChatGPT reached 900 million weekly active users in February 2026, so buyer research now spans several AI interfaces.
- Search Console now reports impressions in AI Overviews and AI Mode, but not the answer text or which competitors were cited.
How is AI changing SEO in 2026?
Five changes matter most for businesses. Each one rests on data or an official statement.
1. More searches end on the results page
When Google shows an AI summary, fewer people click through. A Pew Research Center study tracked 900 US adults and 68,879 Google searches in March 2025.
About 18% of those searches produced an AI summary. Users clicked a traditional result in 8% of those visits, compared with 15% when no summary appeared.
Only 1% of visits included a click on a link inside the summary itself. Users also ended their browsing session more often: 26% of the time with a summary, versus 16% without.
| Pew measure | With an AI summary | Without a summary |
|---|---|---|
| Clicked a traditional result | 8% of visits | 15% of visits |
| Clicked a link inside the summary | 1% of visits | No summary shown |
| Ended the browsing session | 26% of visits | 16% of visits |
This is the core of zero-click search: the question gets answered, and the source may never see the visit.
2. AI Overviews and AI Mode are merging into one Google experience
Google no longer treats AI answers as an experiment on the side. In its Q2 2026 earnings remarks on July 22, 2026, Alphabet said it had brought AI Overviews and AI Mode together into one Search experience.
Sundar Pichai also said AI Mode passed 1 billion monthly active users after its global expansion in October 2025. He added that AI Mode drives an incremental increase in overall Search queries.
For many questions, the AI answer is becoming the default view.
3. AI assistants are now a second discovery channel
Search demand no longer lives only in Google. OpenAI said ChatGPT reached 900 million weekly active users in February 2026, as TechCrunch reported.
In the same Q2 2026 remarks, Alphabet said the Gemini app had 950 million monthly active users. Microsoft Copilot and Claude add further places where buyers ask for advice.
These assistants handle long, specific questions, such as which tool fits a team of ten with a set budget. They often return a shortlist of brands, and what matters is whether your brand makes the list.
4. The unit of visibility is the answer, not the position
In classic search, you rank in a position on a list.
In an AI answer, your brand is mentioned, your page is cited, or you are absent.
Google also uses query fan-out: the model runs several related searches before it writes one answer. A page can be cited for a sub-question you never targeted.
So the useful metrics shift from average position to citation rate, mention rate and share of voice across a set of prompts.
5. Measurement is catching up
Google announced generative AI performance reports in Search Console on June 3, 2026. It rolled them out to all websites by August 31, 2026.
The report shows impressions in AI Overviews and AI Mode by page, country, device and date. It does not show the answer text, your brand mentions or which competitors were cited.
Microsoft added a similar AI Performance report to Bing Webmaster Tools in February 2026. It counts how often your pages are cited in Copilot and Bing AI answers.
| Report | Released | AI surfaces | What it counts |
|---|---|---|---|
| Search Console | June 3, 2026 (all websites by August 31, 2026) | AI Overviews and AI Mode | Impressions by page, country, device and date |
| Bing Webmaster Tools | February 2026 | Copilot and Bing AI answers | How often your pages are cited |
What stays the same: SEO fundamentals still decide eligibility
AI has not replaced SEO. Google’s guide to optimizing for generative AI features says these features are rooted in its core Search ranking and quality systems.
Google describes the process as retrieval-augmented generation. Its ranking systems retrieve relevant pages from the index, and the model writes an answer with links to those pages.
To be eligible, a page must be indexed and allowed to show a snippet. Google also lists a Search Console setting for its generative AI features, and sites are included by default.
The same guide addresses AEO and GEO directly. From Google’s perspective, optimizing for generative AI search is still SEO.
Google also says you can ignore several popular hacks for Google Search:
Google Search doesn’t use llms.txt files, so they neither help nor hurt there.
No special AI schema is required, although structured data still helps with rich results.
There’s no need to chunk pages into tiny pieces for AI.
Seeking inauthentic mentions across the web isn’t as helpful as it might seem.
Other assistants work on the same principle with their own indexes. ChatGPT, Copilot and Claude can only cite pages their search systems can reach and understand.
For a side-by-side view of the three disciplines, read AEO vs SEO vs GEO.
The future of AI search and its impact on organic traffic
The impact depends on the type of query. Two credible sources point in different directions.
Pew’s data shows fewer clicks when an AI summary appears. Google disagrees with the wider conclusion.
In an August 2025 post, Google’s head of Search said total organic click volume was relatively stable year over year. She also said Google sends slightly more quality clicks than a year earlier.
Google did not publish the underlying numbers. It did say traffic is shifting between sites, toward forums, videos and first-hand perspectives.
Both findings can be true at once, because different queries behave differently:
Quick factual questions lose the most clicks, because the answer is complete on the page.
Research, comparison and purchase queries still send visits, because people want depth before they decide.
Brand exposure without a click becomes common, so branded search and direct visits matter more as signals.
Original content such as reviews, data and expert analysis is what Google says people click into to learn more.
Nobody outside Google can see total click data, so treat any precise traffic forecast with care. Measure your own pages instead.
SearchGPT vs Google AI: what differs for business visibility?
SearchGPT was OpenAI’s prototype, and its features are now part of ChatGPT search. It and Google’s AI features answer questions in different settings.
| Area | ChatGPT search | Google AI Overviews and AI Mode |
|---|---|---|
| Setting | ChatGPT search is a conversation. Buyers refine their needs over several turns and often ask for a named shortlist. | Google AI Overviews and AI Mode sit inside the results page, next to organic links, ads, shopping and local results. |
| Sources | ChatGPT search combines OpenAI’s own crawler with third-party search providers. | Google draws on its own index. |
| Main risk | In ChatGPT, the risk is being left off the shortlist. | In Google, it’s a competitor cited above your organic result. |
Track both. A Google-only view misses conversational research. Leave Google out, and you ignore the largest source of search demand.
What our tracking data shows about AI answers
Ansvisor, an open-source AI search intelligence platform, tracks prompts about AI search on six AI platforms. The six are ChatGPT, Google AI Overviews, Google AI Mode, Gemini, Copilot and Perplexity.
From September 15 to October 14, 2026, AI engines cited 746,702 links across 103,427 tracked answers. That is 7.2 links per answer on average.
The spread between platforms was large. Google AI Overviews averaged about 14 links per answer and Gemini about 12. ChatGPT averaged about 5, while Google AI Mode and Microsoft Copilot stayed below 5.
| Platform | Cited links per answer (average) |
|---|---|
| All six platforms | 7.2 |
| Google AI Overviews | About 14 |
| Gemini | About 12 |
| ChatGPT | About 5 |
| Google AI Mode | Below 5 |
| Microsoft Copilot | Below 5 |
Put this next to Pew’s finding that only 1% of visits included a click on a link inside the summary. Being one of many cited sources is common. The click is rare.
That’s why your brand name in the answer text matters. A buyer who reads it may search for you later, and that visit carries no referral you can trace.
What should businesses do now?
You don’t need a new strategy from scratch. Extend the SEO work you already do with these seven steps.
- Check that AI systems can reach your pages. Confirm that robots.txt, your CDN and your firewall allow Googlebot, Bingbot and OAI-SearchBot. Our guide on checking AI crawler access covers the tests.
- Build a prompt set of 25 to 100 buyer questions. Include problem, category, comparison and "best tool for" questions. Your brand name alone tells you little.
- Set a baseline on every platform your buyers use. Record mentions, cited URLs and competitors in ChatGPT, AI Overviews, AI Mode, Gemini, Copilot and Claude.
- Update existing articles before writing new ones. Put a direct answer under each heading, and back it with dates and numbers from named sources. Use our content refresh guide for AI search as a checklist.
- Publish content only you can write. First-hand tests, original data and expert opinions are what Google calls non-commodity content.
- Earn mentions on the sources AI cites. In Ansvisor tracking data, reddit.com, youtube.com, github.com and linkedin.com were the most cited third-party domains.
- Connect visibility to visits and revenue. Segment AI referrals in analytics, as shown in our guide to tracking traffic from ChatGPT.
Rerun the same prompts every month and compare who gets cited and mentioned between runs. Larger organizations can see how these steps scale in our overview of enterprise AI SEO trends.
Do traditional SEO tools still work for AI search?
Classic SEO tools still cover the foundation: crawling, indexing, keyword demand and organic rankings. You need that data, because AI answers draw on search indexes.
Most of them were built to record positions, not answers. They usually cannot show which sources an AI answer cited or how it described your brand.
Search Console and Bing Webmaster Tools now fill part of the gap for Google and Microsoft. Neither covers ChatGPT, Gemini or Claude.
How Ansvisor helps
Ansvisor runs your prompt set on a schedule across the major AI platforms and stores every answer. Answer Engine Insights shows where your brand appears and which competitors lead.
Citation monitoring lists the exact URLs each answer cited, so you know which pages to update first.
You can use Ansvisor Cloud or self-host the open-source project on GitHub.
Start a 14-day free trial and see how AI search describes your business today.
FAQ
Is AI replacing SEO?
No. Google says its AI features are rooted in its core Search ranking and quality systems, and a page must be indexed to be cited. What changes is the goal: you also want to be cited and named inside AI answers.
How does AI affect organic traffic?
It depends on the query. Pew Research found fewer clicks when an AI summary appeared, while Google says total organic clicks were relatively stable in 2025. Quick factual questions lose the most clicks, while research and purchase queries still send visits.
How should brands update existing articles for better AI search performance?
Start with pages that already rank or earn impressions. Put a direct answer under each heading, add specific facts with dates and named sources, and cover the follow-up questions buyers ask. Then check whether AI answers cite the page after two to four weeks.
Can I see AI Overviews and AI Mode data in Search Console?
Yes. Google launched a generative AI performance report in June 2026 and rolled it out to all sites by August 31, 2026. It shows impressions, pages, countries, devices and dates, but not the answer text or which competitors were cited.
Do AEO and GEO replace SEO?
No. Google says that from its perspective, optimizing for generative AI search is still SEO. AEO and GEO are useful names for the extra work of earning citations and mentions in AI answers, from ChatGPT to Copilot.
Cihan Geyik
Co-founder at Ansvisor
Cihan Geyik is the co-founder of Ansvisor, an open-source, cloud-ready AI Visibility platform for AI Search. With more than 15 years of experience in digital marketing and growth, he writes about AI visibility, AI search, AEO, GEO, citations, and answer engines. He focuses on helping brands understand and improve their presence across ChatGPT, Gemini, Perplexity, Google AI Overviews, and other AI-powered discovery platforms.
Get more Ansvisor insights in Google
Add Ansvisor as a Preferred Source to see more of our AI Search, AEO, GEO, and AI Visibility insights across Google.
Make Your Brand Visible in AI Answers.
INDUSTRIES
© 2026 Ansvisor. All rights reserved. Ansvisor is an open-source AI Search Intelligence Platform for AI Visibility. Ansvisor is a brand owned and operated by Empler AI Inc.