Google’s AI Mode Shakes Up Search Landscape With High Volatility and Low Overlap

Google’s AI Mode is already changing how we search online and what we see in the results. 

This new way of searching, powered by artificial intelligence and gives users direct answers, helpful summaries and links to different sources all without following the 

traditional list of blue links we are used to seeing. 

But a new research report by SE Ranking, based on a study of 10,000 keywords, found that AI Mode is not just a small upgrade. 

It is a completely new experience that brings very different results, every time you search.

This change leaves us with more questions than answers. What happens to traditional SEO? This new data gives us some surprising answers.

91% of URL Results Change Across Repeated Searches

According to SE Ranking’s analysis of 10,000 commercial keywords, AI Mode delivers a level of volatility we have never seen in Google Search. Here is what they found. 

Only 9.2% of URLs cited remained consistent across three different sessions of the same query, on the same day.

Domain-level consistency is slightly higher at 14.7% which indicates that while the exact pages change, the domain might remain (e.g., switching from one product page to another on the same site).

In 29.4% of queries, not even a single domain was repeated in all three versions. 

Think about that the search results generated by AI can now shift completely every time a user searches, even if the question does not change.

For SEOs, that flips the playbook. Ranking is a moment in a rotation and we need to understand how to increase our odds of being chosen in that moment. 

Low Overlap With AI Overviews and Organic Rankings

One of the most eye-opening sections of the report deals with how different AIM is from existing Google products.

AIM vs. AI Overviews (AIO):

• URL-level overlap is just 10.7%
• Domain-level overlap is 16%

AIM vs. Organic Search:

• URL overlap with the top 10 organic results is just 14%
• Domain overlap sits at 21.9%

• 17.9% of AI Mode queries had no URLs in common with the organic top 10 and 6.2% had no overlap even in the top 20.

This is huge. AIM does not just add a new layer of logic but essentially ignores what is ranking organically. So being #1 in traditional SEO? Does not guarantee you will be featured in AI Mode at all 

Where Does AI Mode Pull Its Sources From?

One fascinating insight: Google properties dominate AIM responses. That means 5.7% of all citations were to www.google.com. 97.9% of those links pointed to Google Maps business profiles especially when local context was involved. 

And in 9.2% of responses, business cards linking to Google Maps appeared prominently, showing NAP data and reviews. 

In contrast to AIOs, which often send users back to the search page, AI Mode keeps users within the Google ecosystem. It also frequently links to:

• Google Travel (flights, hotels)
• Google Finance
• But rarely links back to organic search results.

This underscores Google’s effort to control the user journey within its AI framework.

Top-Cited Websites in AI Mode

Despite volatility in specific URLs, a few domains consistently emerged as authoritative favorites:

1. indeed.com (1.8%)
2. wikipedia.org (1.6%)
3. reddit.com (1.5%)
4. youtube.com (1.4%)
5. nerdwallet.com (1.2%)

These sites appeared across all datasets, suggesting that AIM favors familiar, trusted domains, even if it rotates which pages are cited

Inline vs Block Links—How Is AI Mode Structuring Sources

Here is how AI Mode uses its 122,617 analyzed citations:

• 90.8% were block links which are placed in a sidebar or below the response.
• 8.9% were in-text links embedded in the response.
• 0.3% were AIM SERP-style links short organic-style sections within the AI response.

Why does this matter?

Because block links are more likely to be skipped by users, while inline links and especially those at the beginning of the response get higher visibility and likely more clicks.

If your content is cited within the first few lines, your CTR could spike. But if you are buried in the sidebar? You might go unnoticed.

Personalization and Location Influence

Even when no location is mentioned in the query, AI Mode appears to adjust results based on the user’s location.

In the dataset: Local queries often triggered business profiles and map links, National queries returned different domains based on searcher region. 

For example:

• A search about “best Thai food” in Texas triggers different results than in Denver.
• Even non-location-based queries return local business profiles depending on where you are.

This personalization hints that Google is now blending semantic AI with local context, even in general queries. It makes sense but it means results vary even more between users. 

So what you see in AIM is probably not what your client, customer or audience sees. 

Search Is Not Dying—It is Becoming AI

Google’s AI Mode isn’t a new tab. It is a new foundation. It moves fast. It is unpredictable. It draws from a much wider universe of sources. And it does not care where you rank in the blue links.

This is a search in 2025 and it rewards intent, depth, context and diversity, not just SEO tactics. The future is clear that the AI search race is already running and visibility now depends on your ability to adapt, not just optimize.

Dileep Thekkethil

Author

Dileep Thekkethil is the Director of Marketing at Stan Ventures, where he applies over 15 years of SEO and digital marketing expertise to drive growth and authority. A former journalist with six years of experience, he combines strategic storytelling with technical know-how to help brands navigate the shift toward AI-driven search and generative engines. Dileep is a strong advocate for Google’s EEAT standards, regularly sharing real-world use cases and scenarios to demystify complex marketing trends. He is an avid gardener of tropical fruits, a motor enthusiast, and a dedicated caretaker of his pair of cockatiels.