The 2026 NAR Technology Report makes one thing pretty clear: AI adoption in real estate is no longer theoretical. The opportunity for brokers is to stop trying to manage the technology and start leading agents toward responsible adoption.
For the last couple of years, I’ve heard brokers ask some version of the same question:
“Should we let our agents use AI?”
I think the 2026 NAR Technology Report officially makes that the wrong question.
Your agents are already using it. As independent contractors, you have limited control over whether they use it, but you can apply guardrails and lead them through best practices.
According to NAR, 23% of REALTORS® now use AI daily. Another 25% use it weekly, and 31% are experimenting with it occasionally. Only 12% say they aren’t using AI and don’t plan to. 2026_technology_report_public.pdfPDF
In other words, AI isn’t sitting outside your brokerage waiting for someone to approve the purchase order.
It’s already inside the building.
And that changes the broker’s role, as leader.
The Number That Got My Attention
There are plenty of interesting statistics in NAR’s report, but one jumped out at me.
Among REALTORS® who use AI, 91% are using ChatGPT. Gemini comes in at 42%, Copilot at 27%, and Claude at 25%.
But I think how agents are using these tools is even more important.
The three most common uses are:
-
Writing listing descriptions: 75%
-
Social media posts: 56%
-
Emails and follow-up: 52%
That probably doesn’t surprise anyone.
But keep reading the list.
Agents are also using AI for market summaries, document review, client education, pricing and comps support, lead generation, transaction recommendations, risk insights, client status reports, contract and disclosure summaries, and even auto-filling forms.
We’ve moved pretty quickly from:
“Write me an Instagram post about this new listing.”
to:
“Read this contract and tell me what my client should do next.”
Those are two very different uses of artificial intelligence.
One helps you come up with a clever caption.
The other can affect somebody’s money, contractual obligations, personal information, negotiations, and potentially their legal rights.
That’s where this becomes a leadership issue.
Brokers Have Two Bad Options
I see two extremes developing around AI.
The first is what I’ll call the Wild West Brokerage.
Agents download whatever AI tool appeared in their Instagram feed that morning, upload whatever documents they want, ask whatever questions they want, and assume that because the answer appeared confidently on the screen, it must be correct.
Then there’s the Fort Knox Brokerage.
AI is scary. Nobody understands it. Compliance has concerns. Therefore, the unofficial policy becomes:
“Please don’t use that stuff.”
Neither approach works.
The Wild West creates unnecessary risk.
Fort Knox creates shadow AI.
Because if nearly half of REALTORS® are already using AI daily or weekly, telling people not to use it isn’t going to stop adoption.
It just means they’ll use it without guidance.
And that’s the difference between managing technology and leading people through technological change.
Managers Ask How to Control It. Leaders Ask How to Use It Well.
That distinction matters.
The NAR report found that 63% of REALTORS® say the learning curve is a challenge when adopting new technology. Another 44% say there are simply too many options and it’s difficult to choose. And 36% cite privacy and security concerns. 2026_technology_report_public.pdfPDF
To me, that sounds less like an agent problem and more like an opportunity for brokerage leadership.
Agents don’t need another 47-page technology policy sitting somewhere in the company intranet next to the 2017 holiday schedule.
They need a safe path to adoption.
Brokers Need to Curate, Not Just Regulate
I believe one of the broker’s emerging responsibilities is becoming a technology curator.
That doesn’t mean chasing every shiny AI object that wanders across YouTube (even if I have built a YouTube channel of shiny objects).
It means identifying where these tools create legitimate value.
NAR found that 81% of REALTORS® adopt technology to save time, while 71% say improving the client experience is a motivation.
Those are pretty good North Stars.
Does this technology make the agent more efficient?
Does it improve the consumer’s experience?
Does it help the agent communicate more clearly?
Does it reduce repetitive work?
Does it allow the agent to spend more time advising clients instead of formatting documents, writing boilerplate emails, or staring at a blinking cursor?
If the answer is yes, brokers should be exploring it.
But there’s another question sitting right beside those:
Can we use it responsibly?
Because efficiency without safeguards isn’t innovation.
Sometimes it’s just faster risk.
We Need an AI Playbook, Not an AI Prohibition
This is where I think brokerages need to mature quickly.
Instead of issuing vague warnings about AI, brokers should establish a practical framework for using it.
Agents should know which tools the brokerage has evaluated and recommends.
They should understand what information can and cannot be uploaded.
They should know when AI-generated work requires human verification.
They should understand the risks surrounding confidential information, Fair Housing, copyright, hallucinations, advertising, contracts, disclosures, and client communications.
And most importantly, they should understand a very simple principle:
AI can assist professional judgment. It cannot replace professional responsibility.
That distinction is going to become increasingly important as AI moves deeper into the transaction.
The report already shows agents using AI for pricing support, document review, client education, transaction recommendations, risk insights, and contract or disclosure summaries.
We’re not just talking about marketing anymore.
We’re getting closer to the machinery of the transaction.
That’s where brokers need to be standing.
Not blocking the doorway.
Not pretending the doorway doesn’t exist.
Standing beside their agents and showing them how to walk through it safely.
Consumer Protection Is Part of AI Adoption
There’s another reason brokers need to lead this.
Consumers generally seem receptive to technology.
In NAR’s survey, 40% of REALTORS® said clients responded very positively to technology in the buying and selling process. Another 37% reported positive reactions, although those consumers had some reservations.
That’s 77% reporting some level of positive consumer response.
Consumers aren’t asking us to become less technologically capable.
They’re expecting us to become technologically capable without becoming professionally careless.
That’s the bargain.
Use AI to make the experience faster.
Use it to communicate better.
Use it to explain complicated concepts.
Use it to organize information.
Use it to reduce administrative work.
But don’t outsource accountability to the robot.
Your client didn’t hire ChatGPT. They hired you.
And when an agent operates under a brokerage, the broker has a responsibility to create an environment where innovation and consumer protection can coexist.
The Real Risk Isn’t AI
I don’t believe the biggest AI risk facing brokerages is that agents will use it.
The data suggests that ship has already left the dock.
The bigger risk is that agents will use increasingly sophisticated AI tools without training, standards, supervision, or a clear understanding of where the boundaries are.
NAR found that 67% of REALTORS® say they use some emerging technologies but are still learning. Only 12% say they use them well enough that they could teach others.
That gap matters.
Because somewhere between:
“I’m experimenting with this.”
and
“I’m relying on this to help advise my client.”
sits the broker.
And that’s where leadership belongs.
Don’t Manage the Future. Lead People Into It.
Every major technological shift in real estate has created some version of this conversation.
Websites. Electronic signatures. Social media. Smartphones. Cloud transaction management. Virtual tours.
Now AI.
The technology changes.
The broker’s responsibility doesn’t.
We identify what creates value.
We understand the risks.
We establish reasonable guardrails.
We educate our agents.
We protect consumers.
And then we give good professionals enough room to innovate.
The brokerages that handle AI well won’t necessarily be the ones with the longest policies or the biggest technology budgets.
They’ll be the ones whose leaders can look at their agents and say:
“We’ve looked at this. We understand where it can help you. We understand where it can hurt you. Here’s what we recommend. Here’s what we don’t. Here’s how we’re going to train you. And here’s how we’re going to protect the client while we do it.”
That’s leadership.
And right now, our agents need a lot more of that than they need another memo.
For a list of AI tools, resources, risk mitigation resources and tutorials, visit my AI toolkit at www.matthewrathbun.com/ai
Source: National Association of REALTORS®, 2026 REALTORS® Technology Report.
This article is also part of Here, Let Me Help, my newsletter on practical systems, AI, leadership, and running a real estate business. Want the next one in your inbox? Subscribe here.