Are Homebuyers Using AI to Find Realtors? What the Research Says

Yes—but the evidence needs to be interpreted carefully.

There is now strong evidence that homebuyers and sellers are using AI for housing research, neighborhood research, affordability questions, property discovery and local business recommendations.

There is also growing direct evidence that consumers are asking AI systems such as ChatGPT and Gemini to recommend real estate agents.

What we do not yet have is a definitive nationwide statistic saying:

“X% of homebuyers found their Realtor through ChatGPT.”

The market is still too new for that kind of standardized measurement.

But multiple 2025 and 2026 studies from organizations including the National Association of Realtors, Realtor.com, Bank of America, CMHC, McKinsey, Cotality, BrightLocal and independent AI visibility researchers point in the same direction:

AI is becoming part of the early real estate discovery journey—and that increasingly includes researching which Realtor to work with.

Are Homebuyers Actually Using AI?

Yes.

This is now well supported by multiple large consumer studies.

A 2025 Realtor.com survey of 1,000 U.S. adults who had recently bought or sold a primary residence, or were considering doing so, found that:

82% were using AI for real estate market information.

Among respondents:

• 67% used ChatGPT
• 54% used Gemini

When asked which information sources made them smarter about the housing market, real estate agents ranked first at 62%, while AI followed closely at 61%.

This does not mean 82% used AI to choose a Realtor.

But it does show that AI has already become a mainstream source of real estate information among active and recent buyers and sellers.

That matters because agent discovery happens inside the same research journey.

Bank of America: 1 in 5 Are Already Using AI for Homebuying Research

Bank of America's 2026 Homebuyer Insights Report provides another strong signal.

The report found that:

20% of prospective buyers and current homeowners used AI tools or chatbots for homebuying research during the previous year.

Usage was higher among younger generations:

• Millennials: 28%
• Gen Z: 32%

Among respondents who used AI:

• 57% used it to estimate affordability, mortgage payments or closing costs
• 55% used it for education and research about the homebuying process
• 52% used it to research neighborhoods, market trends or property values.

The important point is not simply that people are “using ChatGPT.”

They are using AI for questions that sit directly inside the homebuying funnel.

They are researching:

Can I afford this?

Where should I live?

What is this neighborhood like?

What is this property worth?

The next logical question in that journey can easily become:

Who should help me buy it?

The Same Trend Is Appearing in Canada

This is not limited to the United States.

The Canada Mortgage and Housing Corporation's 2026 Mortgage Consumer Survey found that:

16% of consumers who searched online for mortgage information used AI to obtain mortgage-related information.

For Canadian Realtors, this matters because mortgage research is closely connected to homebuying intent.

A consumer using AI to understand:

• mortgage qualification
• affordability
• down payments
• closing costs

is already deep inside the housing decision process.

AI is increasingly present before the consumer reaches the transaction stage.

McKinsey Says AI Is Changing Housing Discovery

McKinsey's July 2026 analysis of residential real estate goes further.

McKinsey says AI is changing how consumers:

• discover housing
• compare options
• get questions answered
• assess affordability
• schedule tours
• build trust.

McKinsey also notes that while traditional search engines, portals, broker websites and property websites still matter, consumers are increasingly incorporating:

AI-powered answer engines, AI assistants and recommendation-driven discovery

into the front end of the housing journey.

That distinction is important.

AI is no longer only something real estate companies use internally.

It is increasingly something consumers use before they decide which company, property or professional to contact.

Are Buyers Specifically Asking AI to Find Realtors?

This is where the evidence becomes more direct.

In May 2026, the National Association of Realtors published guidance specifically about GEO and AI-driven agent discovery.

NAR stated that buyers and sellers are increasingly using tools such as:

• ChatGPT
• Gemini
• Claude
• Perplexity

to ask questions including:

“Who is the best real estate agent near me?”

and:

“Who is the top listing agent in my area?”

NAR's conclusion was straightforward:

AI is becoming a new pathway connecting consumers to real estate professionals.

This is one of the strongest current pieces of direct evidence that Realtor discovery is moving into AI interfaces.

NAR Has Already Documented a Real Client Coming From Gemini

NAR also documented a concrete example.

According to its May 2026 article, Denver luxury Realtor Rick Janson said a prospective seller found him through a Gemini search for Denver real estate agents.

The AI recommendation reportedly summarized his credentials, video content and professional background before the client ever contacted him.

According to the Realtor, that AI-originated discovery led to a $1.7 million listing appointment.

This is an anecdote—not population-level evidence.

It should not be interpreted as proof that Gemini is already a major Realtor lead source across the industry.

But it proves the behavior is no longer theoretical.

A consumer can:

ask AI for a Realtor → receive a recommendation → contact that Realtor → become a real lead.

NAR Now Calls AI an Early Step in the Homebuying Journey

NAR reinforced the trend again in June 2026.

Its article “AI Becomes Early Step in Homebuying Journey” summarized Bank of America and Cotality consumer research and described AI as increasingly important in the informational phase of homebuying.

NAR also specifically advised real estate professionals to pay attention to how they appear across AI-driven platforms.

That is an important change in industry language.

The question is no longer:

“Will buyers ever use AI?”

It is becoming:

“Where in the homebuying journey are they already using it?”

Cotality: 55% of Homebuyers Use Generative AI at Least Monthly

Cotality's 2026 international housing research adds another layer.

Its survey of recent and prospective homebuyers across the United States, Canada, the United Kingdom and Australia found that:

55% of homebuyers use generative AI tools at least once a month.

It also found that 75% of buyers assume AI is already being used somewhere inside the homebuying process.

This is important because it shows AI is becoming normalized in housing.

However, Cotality also found something equally important:

Consumers still want human verification.

Among surveyed buyers:

• 68% said they would manually verify all or a significant amount of AI-provided housing information
• 44% said they would pay more to have a human professional verify AI-generated housing decisions.

This suggests AI and Realtors are not necessarily competing for the same role.

A more likely model is:

AI helps consumers research. Realtors help consumers verify, decide and transact.

AI Does Not Appear to Be Replacing Realtors

This point matters.

The available evidence does not show that buyers are abandoning real estate professionals in favor of AI.

Bank of America found that even consumers who use AI still prefer human expertise for important parts of the transaction.

Among prospective buyers:

• 55% preferred human expertise for touring homes
• 54% preferred humans for legal or contractual advice.

Cotality found a similar preference for human oversight.

And Realtor.com's survey found real estate agents remained the most trusted and accurate housing information source among respondents, even as AI adoption increased.

So the important change is not:

AI replacing Realtors.

It is:

AI influencing which Realtors consumers discover before they ever speak to one.

Consumers Are Already Using AI for Local Business Recommendations

There is also broader consumer evidence beyond real estate.

BrightLocal's 2026 research found that:

45% of consumers had used AI tools for local business recommendations during the previous year.

The previous year's figure was only 6%.

Among active AI users:

63% trusted AI business recommendations.

BrightLocal also found that 42% of consumers trusted AI recommendations as much as traditional online reviews.

This matters for Realtors because real estate agents are fundamentally local professional service providers.

Someone searching:

“Best Italian restaurant near me”

and someone asking:

“Best Realtor for a first-time buyer near me”

are different transactions.

But they share the same underlying discovery behavior:

Ask AI to reduce a large local market into a small recommendation set.

BrightLocal Says AI Is Becoming a Major Local Discovery Channel

BrightLocal's broader 2026 consumer research found that AI had rapidly moved into the mainstream for local search.

Its reporting describes AI tools such as ChatGPT as becoming one of the most common places consumers look for local recommendations.

This does not prove every category behaves identically.

Choosing a Realtor involves more risk than choosing a restaurant or plumber.

But it does show that consumer behavior is changing.

People are learning to ask AI:

“Who should I use?”

instead of only:

“Show me websites about this.”

What Happens When Consumers Ask AI for Realtors?

This is where independent AI visibility research becomes useful.

These studies do not measure consumer adoption.

Instead, they test what happens after someone asks AI for a Realtor.

Local Falcon ran 37,500 Realtor-related AI searches across the 100 largest U.S. cities in 2026.

It tested platforms including:

• ChatGPT
• Gemini
• Google AI Mode
• Google AI Overviews
• Grok.

Among the established agents and teams benchmarked:

91.5% were never cited once by the AI systems tested.

This suggests that when AI does recommend agents, visibility may be highly concentrated.

In practical terms:

AI does not show buyers every Realtor in the market.

It creates a shortlist.

Another Study Found AI Mentioned Only 1.3% of Realtors in the Markets Tested

withNotable conducted another Realtor-specific AI recommendation study in spring 2026.

It ran:

• 240 queries
• across 10 U.S. markets
• using ChatGPT, Perplexity and Gemini.

The study identified 656 Realtors across the results.

withNotable estimated there were roughly 51,870 licensed professionals across those markets, meaning AI mentioned only about:

1.3% of available Realtors.

The study also found AI usually recommended only:

3–5 Realtors per market.

This supports an important distinction between Google and AI.

Google is often a ranking environment.

AI is increasingly a shortlist environment.

AI Is Actively Reading Realtor and Brokerage Profiles

Another form of evidence comes from server logs.

Renowned analyzed AI bot activity on RateMyAgent during May and June 2026.

It identified approximately:

249,720 page fetches by AI systems while answering live consumer questions.

Those fetches included:

• 126,000 brokerage pages
• 57,000 agent profiles
• 20,000 location pages
• 5,500 property listing pages.

This does not mean 249,720 consumers were asking specifically for Realtor recommendations.

Some questions may have concerned properties, brokerages, reviews or general real estate information.

But it provides direct technical evidence that AI systems are actively retrieving agent and brokerage reputation data while producing real-world answers.

AI systems are not only reading generic housing articles.

They are reading pages about specific agents and brokerages.

Reviews Appear to Be Part of the AI Research Process

The RateMyAgent server-log research found another interesting pattern.

Agents with larger review profiles were fetched more frequently by AI systems.

Average AI fetches per profile increased from:

• 1.8 for agents with no reviews
• 4.6 for agents with 101–250 reviews
• 13.5 for agents with 500+ reviews.

This is correlation, not proof of a ranking factor.

We cannot conclude:

“Get 500 reviews and ChatGPT will recommend you.”

There are many possible reasons high-review agents were read more frequently.

But the finding supports a broader idea:

Reputation content is part of the information AI systems can use when researching Realtors.

Top-Producing Realtors Are Not Automatically AI-Visible

Another surprising result from Local Falcon's study:

Only 21.2% of the highest-volume individual agents in the markets it studied were cited by any of the AI platforms tested.

This matters because consumers may assume:

“AI will simply recommend the most successful Realtor.”

Current testing suggests it is not that simple.

AI visibility can depend on:

• whether the agent is clearly represented online
• what sources mention them
• whether their expertise is understandable
• whether their business information is consistent
• how relevant they are to the specific question
• which AI platform is answering

Real-world production and AI visibility are not identical.

Different AI Platforms Recommend Different Realtors

There is another important finding.

There is no universal AI Realtor ranking.

Local Falcon found substantial disagreement between AI systems when recommending agents.

withNotable also found different results across ChatGPT, Gemini and Perplexity.

BrightLocal's much larger local-business study reached a similar conclusion.

It analyzed 200,875 local AI searches and approximately 1.9 million citations across ChatGPT, Google AI Mode and Google AI Overviews.

One of its major findings was that traditional Google Maps visibility did not translate directly into AI visibility.

Google Maps mentioned the tracked business in about:

66% of searches

while AI surfaces mentioned it in only:

32–38% of searches.

This means Realtors should not think about AI visibility as one ranking.

A Realtor can appear in:

Gemini but not ChatGPT.

Or:

AI Mode but not AI Overviews.

Visibility needs to be measured across platforms.

AI Recommendations Are Usually Very Short

BrightLocal's research also found that AI systems usually recommend only a small number of local businesses.

Across the platforms studied, average recommendation sets were roughly:

2–4 businesses per answer.

withNotable's Realtor-specific research similarly found that AI commonly returned only 3–5 Realtors.

This matters enormously for Realtors.

A traditional Google search can expose a consumer to:

10 organic results,

multiple map listings,

directories,

ads,

and additional pages.

AI can compress that entire discovery process into several names.

The strategic question changes from:

“Can I appear somewhere on page one?”

to:

“Can I make the shortlist?”

So Are People Actually Finding Realtors Through AI?

The most accurate answer today is:

Yes, some clearly are.

NAR has documented buyers and sellers using AI to ask for agent recommendations.

It has also documented at least one real listing opportunity that originated from a Gemini Realtor search.

Independent research demonstrates that AI systems actively return Realtor shortlists when asked.

Server-log data shows AI systems retrieving real agent and brokerage profiles while answering consumer questions.

And broader consumer research shows people are rapidly adopting AI for both housing information and local business recommendations.

What we cannot yet say is:

“AI is already the primary way most buyers find their Realtor.”

There is not enough high-quality evidence for that claim.

What We Know vs What We Don't Know

What the evidence currently supports

We know that:

• AI use for housing research is already substantial.
• Younger buyers are especially likely to use AI.
• Consumers are using AI for affordability, neighborhood and market research.
• Consumers increasingly use AI for local business recommendations.
• NAR says buyers and sellers are already asking AI which agents to hire.
• AI systems can return very small Realtor shortlists.
• Different AI platforms recommend different agents.
• AI systems actively retrieve Realtor, brokerage and review information.
• Strong sales production does not guarantee AI visibility.

What the evidence does not yet support

We do not yet know:

• exactly what percentage of homebuyers choose their final Realtor because of AI
• whether ChatGPT generates more Realtor leads than Gemini
• how AI-originated Realtor leads convert compared with Google or referrals
• whether AI recommendation volume is already material for the average Realtor
• the exact ranking formulas used by AI systems

Those are important distinctions.

Good GEO strategy should be based on evidence, not hype.

Why This Matters Even Before AI Becomes a Major Lead Channel

A Realtor might reasonably ask:

“If only a small percentage of my leads come from AI today, why should I care?”

Because discovery behavior can change before attribution systems catch up.

Consider how consumers already use AI.

A buyer may:

1. Ask ChatGPT which neighborhoods to consider.
2. Ask about property types and prices.
3. Ask which Realtors specialize in those neighborhoods.
4. Search those names on Google.
5. Read reviews.
6. Visit Instagram.
7. Contact the Realtor directly.

The Realtor may record that lead as:

Google

or:

Website

or even:

Direct

when the original discovery source was AI.

AI can influence consideration without receiving final-click attribution.

That makes AI visibility potentially important before it becomes easy to measure in traditional CRM systems.

AI May Become a New Top-of-Funnel Layer

McKinsey's description of the housing journey helps explain this.

AI is increasingly entering the process at the discovery and comparison stage.

That means the future customer journey may look like:

Ask AI

↓

Receive several options

↓

Verify on Google, reviews and social media

↓

Visit Realtor website

↓

Contact agent

AI does not necessarily replace existing channels.

It can sit above them.

This is why AI visibility is better understood as a consideration metric, not just a traffic metric.

Realtors Still Have an Advantage AI Cannot Replace

There is another conclusion across almost all of the consumer research:

Consumers still value human expertise.

Realtor.com found agents remained the most trusted housing information source.

Bank of America found buyers still preferred human professionals for tours and legal or contractual guidance.

Cotality found that 44% of buyers would even pay more for a human expert to verify AI-generated housing decisions.

So the opportunity for Realtors is not to compete with AI.

It is to become one of the professionals AI helps consumers discover.

What Should Realtors Measure?

Because there is not yet a standardized “AI lead share” metric, Realtors can start with visibility.

Useful questions include:

• Does ChatGPT recommend me?
• Does Gemini recommend me?
• Do I appear in Google AI Mode?
• Do I appear in AI Overviews?
• Which buyer questions trigger my name?
• Which seller questions trigger my name?
• Which competitors appear instead?
• What websites does AI cite?
• How often do I appear across repeated searches?
• What specialties does AI associate with me?

Those questions reveal something traditional SEO tools often cannot:

How AI currently understands your position in the market.

Frequently Asked Questions

Are homebuyers using ChatGPT to buy homes?

Yes, consumers are already using ChatGPT and other AI systems for housing research.

Realtor.com's 2025 consumer survey found 82% of its surveyed recent or prospective buyers and sellers used AI for real estate information, with ChatGPT the most commonly used platform at 67%.

Are buyers using AI to find Realtors?

There is growing direct evidence that they are.

NAR reports that buyers and sellers are increasingly asking AI tools questions such as “Who is the best real estate agent near me?” and has documented an agent receiving a listing opportunity after being discovered through Gemini.

However, there is not yet a definitive national percentage measuring how many buyers ultimately select their Realtor through AI.

How many Realtors does AI recommend?

There is no fixed number.

withNotable found AI frequently recommended approximately 3–5 Realtors in its 2026 testing across 10 U.S. markets.

Broader BrightLocal local-business research also found AI recommendation sets tend to be small, averaging roughly 2–4 businesses depending on the platform.

Does ChatGPT recommend the highest-producing Realtors?

Not necessarily.

Local Falcon found that only 21.2% of the highest-volume individual agents in the markets it studied appeared in any of the AI results tested.

Are younger buyers more likely to use AI?

Yes.

Bank of America's 2026 report found AI homebuying research usage at:

• 28% among Millennials
• 32% among Gen Z.

Is AI replacing real estate agents?

Current research does not support that conclusion.

Consumers are using AI heavily for research, but they still prefer human professionals for important high-stakes parts of the transaction.

The Bottom Line

The evidence does not show that AI has replaced Google, referrals or traditional real estate discovery.

But it clearly shows something new is happening.

Consumers are already using AI to:

• understand housing markets
• calculate affordability
• research neighborhoods
• compare options
• discover local businesses
• and increasingly ask which real estate professionals they should work with.

NAR is already documenting AI-originated Realtor discovery.

McKinsey says AI is moving into the front end of housing discovery.

Realtor.com finds widespread AI use among active buyers and sellers.

Bank of America and CMHC show AI becoming part of homebuying and mortgage research.

BrightLocal shows consumers becoming comfortable asking AI for local business recommendations.

And Realtor-specific studies show AI reducing thousands of agents into very small recommendation sets.

The question for Realtors is therefore no longer only:

“Are buyers using AI?”

They are.

The more important question is:

“When they use AI to find someone like me, do I appear?”

Maahes currently runs your selected local questions through the Gemini API each week. You can see whether you appear, how you are described, and which sources are cited when available. This does not test the consumer Gemini app or other platforms discussed here.

Start with an AI Visibility Test.