Author: Coach Emily

What Does an AI Keynote for Real Estate Agents Cost?

Written by Coach Emily on September 28, 2026. Posted in Uncategorized.

By Emily Terrell — Top Coach and Speaker at Tom Ferry International. Active San Antonio agent closing 70+ transactions a year.

An AI keynote for real estate agents typically costs $2,500 to $15,000 before travel, based on 2026 fee data from Talkadot and National Speakers Bureau. Where a speaker lands depends on domain specificity, format, and whether the session includes a live build. This guide covers the benchmarks, what the fee includes, and how to budget for results.

Key Takeaways

Most working professional speakers charge $2,500 to $15,000 for a keynote before travel, according to 2026 data from Talkadot and National Speakers Bureau.
AI has the widest fee spread of any speaking topic, and real estate domain experience is what separates the tiers.
Travel, recording rights, and bureau commissions can add thousands to a quote, so compare all-in numbers, not headline fees.
A keynote shifts what agents believe and a training changes what they do, so pick the format before you compare prices.

What is an AI keynote for real estate agents?

An AI keynote for real estate agents is a 45- to 60-minute stage session that shows agents where AI fits in their business: listing marketing, follow-up, content, and compliance. It uses real estate workflows instead of generic tech examples. It’s built to shift beliefs and create urgency. It isn’t hands-on training, which typically runs a half day and has agents build on their own listings.

Why does the fee matter for brokerages and event planners?

The fee either buys an outcome or buys an hour, and most events never decide which one they’re paying for. According to NAR’s 2025 Technology Survey (September 2025), 17% of agents reported AI had a significantly positive impact on their business. Another 46% said it had no noticeable impact at all. The same survey found 20% of agents use AI daily and 22% weekly.

Agents already have the tools. What they’re missing is the actual how.

“Technology continues to be a powerful force in real estate, driving efficiency and marketing innovation.”
— Jessica Lautz, Deputy Chief Economist, National Association of REALTORS®

That’s the gap a speaker fee should be judged against. The test isn’t whether the room felt energized. It’s how many agents are running a new workflow 30 days later. A speaker who can’t close that gap is expensive at any price.

How much does an AI keynote cost in 2026?

Here’s how the two most-cited 2026 benchmarks line up:

Tier	Fee range (before travel)	Source
Market-wide median	$2,500	Talkadot, 2026
Association median	$3,215	Talkadot, 2026
Corporate median	$4,500	Talkadot, 2026
Mid-range professional	$5,000–$15,000	National Speakers Bureau, 2026
Established professional	$15,000–$30,000	National Speakers Bureau, 2026

What’s the market benchmark for keynote speaker fees?

The market-wide median keynote fee is $2,500. According to Talkadot’s 2026 keynote fee benchmarks (June 2026), the 75th percentile sits at $5,000 and the 90th at $10,000. Segment medians matter more than the overall number: associations land at $3,215 and corporate teams at $4,500. The dataset draws on more than a million verified audience survey responses collected from January 2023 through March 2026.

How do speaker bureaus tier their fees?

Bureaus quote higher bands than market-wide data. National Speakers Bureau’s 2026 cost guide places:

entry-level speakers at $1,500 to $5,000
mid-range speakers at $5,000 to $15,000
professional speakers at $15,000 to $30,000
celebrity speakers at $30,000 and up

The gap reflects what each source measures. Talkadot reports what the full market logs. Bureaus price the rosters they represent.

Why do AI speaker fees vary more than any other topic?

The variable is domain specificity. Talkadot found the AI category has the widest fee spread of any topic, roughly $15,000 between the 25th and 90th percentile. It names industry-specific applications as the factor that separates the tiers.

A generalist AI futurist and a practitioner who builds real estate workflows live on stage carry the same topic label. They’re priced very differently. For a room full of agents, the second one is the speaker whose content actually gets implemented.

What does a speaker fee include, and what’s billed separately?

The fee almost always covers the talk and one planning call. Nearly everything else is a separate line item. Talkadot’s 2026 guide lists the common add-ons:

Travel, hotel, and ground transportation, which it estimates at $500 to $3,000 for domestic events
Video recording rights, typically $500 to $2,500
Custom content development beyond the speaker’s existing material
A workshop or breakout session after the keynote

Three pricing levers change the total:

Format: Virtual keynotes typically run 20 to 40 percent below in-person fees.
Timing: Bookings made inside 30 days of the event often carry a 10 to 30 percent rush premium.
Channel: Talkadot cofounder Arel Moodie reports planners see bureau commissions of 20 to 30 percent built into the invoice. National Speakers Bureau states most speakers charge the same rate whether booked direct or through a bureau. Ask how the commission is handled before you sign.

Should you pay for a keynote or a training?

Pay for a keynote when the goal is belief and urgency. Pay for training when the goal is adoption. A keynote can convince a room that AI matters. It can’t install a workflow in 45 minutes.

The full agenda for a session that does is in AI training for real estate agents: what it must cover.

The price gap is often smaller than planners expect. Talkadot’s 2026 data shows a keynote-plus-workshop bundle frequently lands within 10 to 20 percent of the keynote-only price. Training engagements also rebook at 2.6 events per relationship on average, compared with 2.1 for workshops. That’s a signal buyers see repeat value in formats that change behavior.

“A speaker fee is a cost until you define the behavior you’re buying. Name one workflow and a 30-day number before you sign, and the fee becomes an investment you can measure.”
— Emily Terrell, Tom Ferry Coach

Common mistakes

Comparing headline fees instead of all-in costs. A $6,000 quote with travel included can cost less than a $5,000 quote with airfare, hotel, and recording rights billed on top.
Booking a generic AI speaker for a real estate room. Agents tune out examples that don’t touch listings, leads, or their CRM. Domain experience is what they’ll implement.
Paying keynote rates for a training goal. If the goal is adoption and nobody opens a laptop, you bought inspiration and expected implementation.
Booking inside 30 days. Rush premiums and compressed preparation cost money. For AI sessions, they also cost the pre-work that makes a live build possible.
Signing without a success metric. Decide what “worked” means before the contract: one named workflow and a 30-day adoption count.

Frequently Asked Questions

How much does an AI keynote speaker cost for a real estate event?

Most working professional speakers charge $2,500 to $15,000 for a keynote before travel, based on 2026 benchmarks from Talkadot and National Speakers Bureau. Talkadot’s data puts the association median at $3,215 and the corporate median at $4,500. AI speakers show the widest spread of any topic, so domain experience in real estate matters more than the AI label.

Does the speaker fee include travel?

Usually not. Most keynote fees cover the talk and a pre-event call, with airfare, hotel, and ground transportation billed separately. Talkadot estimates domestic travel adds $500 to $3,000. Recording rights, custom content, and a follow-up workshop are also commonly priced on top, so ask every speaker for an all-in number before comparing quotes.

Is a virtual AI keynote cheaper than in-person?

Yes. Virtual keynotes typically run 20 to 40 percent below in-person fees, according to Talkadot’s 2026 benchmarks, and they remove travel costs entirely. The tradeoff for AI sessions is the live build: it’s harder to coach agents through their own listings on a screen, so virtual works best for awareness sessions, not adoption.

Do speaker bureaus charge event planners extra?

Sometimes, and the sources disagree. Talkadot cofounder Arel Moodie reports bureau commissions commonly run 20 to 30 percent of the speaker fee and appear in the gross total. National Speakers Bureau states most speakers charge the same rate whether booked direct or through a bureau. Ask any bureau how the commission is handled before you sign.

How far in advance should you book an AI keynote speaker?

Book at least four months out for a mid-tier speaker. Talkadot’s 2026 benchmarks show bookings inside 30 days typically add a 10 to 30 percent rush premium, driven by last-minute travel and compressed preparation. For an AI session with a live build, the extra time also lets the host enforce pre-work so agents arrive with real listings ready.

What’s the difference between paying for an AI keynote and an AI training?

A keynote changes what agents believe; training changes what they do. Keynotes run 45 to 60 minutes and cost less. Training runs a half day or longer, includes a live build on agents’ own listings, and typically costs more. Talkadot’s data shows training engagements rebook at 2.6 events per relationship, a sign buyers see repeat value.

Can two brokerages split the cost of a speaker?

Yes. The National Speakers Bureau recommends exploring partnerships with associations or sponsors to share speaker costs. Talkadot’s 2026 guide adds that committing to two or three events in the same year gives planners real negotiating power, with later dates often priced 20 to 40 percent below the first. Co-hosting across offices or a regional association captures both.

Bring this to your team or event

Emily Terrell speaks at brokerage events, real estate conferences, and team trainings on AI, systems, and social media — the exact playbook in this post, delivered live to your audience. As a Top Coach and Speaker at Tom Ferry International and an active agent closing 70+ transactions a year, Emily speaks from the stage about what’s working right now, not theory. Recent stages include NAHREP and eXp Con.

Book Emily to speak at your next event:
Email: eterrell@yourcoach.com
Phone: (210) 400-9191
Web: coachemilyterrell.com

For real estate agents who want to implement this: Get the weekly real estate prompt library at weeklyrealestateprompts.com or follow @coachemilyterrell on Instagram for daily systems and AI breakdowns.

Is It Safe to Put Client Data in ChatGPT? 2026 Rules

Written by Coach Emily on September 21, 2026. Posted in Uncategorized.

By Emily Terrell — Top Coach and Speaker at Tom Ferry International. Licensed since 2016. Closing 70+ deals/year while coaching agents nationwide.

Putting client data in ChatGPT is safe only on accounts that exclude your inputs from model training — ChatGPT Business, Enterprise, Edu, or the API. Personal Free, Go, Plus, and Pro accounts train on your conversations by default. This guide covers the exact account settings, the redaction rule, and the brokerage policy questions to ask.

Key Takeaways

Safety is a property of your account tier, not of ChatGPT as a product — the same prompt is compliant on one plan and a disclosure risk on another.
Paying for Plus or Pro buys capability, not privacy. Business, Enterprise, Edu, and API accounts are the tiers excluded from training by default.
A retention policy is a promise a company makes, and a court can override it — the 2025 New York Times litigation proved that in writing.
Redaction beats permission. Strip names, addresses, contact details, and financial specifics and you get the same output with none of the exposure.
Your biggest risk isn’t you. It’s a team member or VA pasting a full CRM export into a free account to clean it up.

What is client data in an AI tool?

Client data is any information a buyer, seller, tenant, or landlord gave you because you’re their agent — names, addresses, phone numbers, email addresses, loan pre-approval amounts, bank statements, divorce or relocation circumstances, and the single most valuable item in the file: motivation. When you paste that into a chat window, you’re transmitting it to a third-party vendor. The question is never whether the vendor is trustworthy. It’s what your contract with that vendor actually says, and whether you have one at all.

There’s a second category agents forget: data that isn’t confidential on its own but becomes identifying in combination. “Seller on Bandera Road” plus “needs to close before the school year” plus “divorce” is a named person to anyone in your market, even with the name stripped.

Why this matters for real estate agents

Volume is the problem. According to NAR’s 2025 Member Profile (August 2025), the typical Realtor completed 10 transaction sides in 2024 with a median sales volume of $2.5 million. That’s ten files of financial and personal information moving through one agent’s hands in a year — and the average agent is now running some portion of that through an AI tool for descriptions, follow-up, CMAs, and marketing copy.

Confidentiality isn’t a preference you can trade for speed. Under the REALTOR® Code of Ethics, the obligation to preserve the confidentiality of client information continues after the transaction closes. Risk guidance for brokerages is direct on this point: don’t put confidential, proprietary, or personal client information belonging to your client, your brokerage, or anyone else into a generative AI platform, because the terms of use on most consumer platforms permit the platform to use what you provide.

NAR’s own legal team has flagged three related traps — accuracy, copyright, and scope. Chloe Hecht, NAR senior counsel, warns that AI platforms are not fully accurate, which makes agent oversight critical, and cautions against using AI to draft contracts, modify standard forms, or give legal advice, since Article 13 of the Code and many state statutes prohibit the unauthorized practice of law.

“Your brokerage doesn’t need an AI policy because AI is risky. It needs one because one team member pasting a CRM export into a free account creates a disclosure you can’t take back.” — Emily Terrell, Tom Ferry Coach

The three-layer safety check

Run these in order. If layer one fails, nothing below it matters.

Which ChatGPT account are you actually typing into?

This is the whole ballgame, and most agents have never checked. OpenAI’s documentation states that by default it does not train on inputs or outputs from business products, including ChatGPT Business, ChatGPT Enterprise, and the API, and that organizations are opted out of data sharing unless they explicitly opt in. Personal accounts work differently: on Free, Go, Plus, and Pro plans, conversations may be used to improve future models unless you turn that off yourself.

The setting lives in Settings → Data Controls → “Improve the model for everyone.” Turn it off. Then understand what that does and doesn’t fix: it’s forward-looking only. Anything already included in a completed training run cannot be pulled back out.

If you have an assistant, a transaction coordinator, or a VA, check their account too. Yours being configured correctly does nothing for the one they opened on a personal Gmail in 2024.

What does the platform keep, and for how long?

Here’s the part nobody in real estate talks about. In May 2025, a federal magistrate judge in the New York Times copyright case ordered OpenAI to preserve and segregate output log data that would otherwise have been deleted — overriding the company’s own 30-day deletion policy for consumer accounts. Chats users believed they had deleted were still there.

OpenAI has since confirmed that obligation ended on September 26, 2025, and that deleted conversations and Temporary Chats are again removed within 30 days, with a limited set of April–September 2025 data still held under legal hold. Enterprise and Edu accounts were carved out of the preservation order throughout. In late 2025 the court went further and ordered production of 20 million de-identified chat logs to the plaintiffs.

The lesson isn’t that OpenAI behaved badly — the company fought the order publicly and largely won. The lesson is structural, and it applies to every vendor you’ll ever use: a retention policy is a promise, and a court can vacate it overnight without consulting you or your client.

What do your brokerage and your license require?

Your platform settings are the floor, not the ceiling. NAR now publishes a brokerage AI policy template covering client and consumer data, MLS content, CRM, transaction documents, and valuations — which tells you that brokerages are actively writing these rules right now, and that a brokerage without guardrails is treating AI as a compliance exposure rather than a productivity tool.

Ask your broker three questions this week: Which AI tools are approved? Who owns the workspace settings? What categories of data are prohibited? If nobody can answer, you’ve found a project — and, if you’re a team leader, a liability sitting on your desk.

This is general information, not legal advice. Confirm your specific obligations with your broker and a licensed attorney in your state.

The redaction rule: what to strip before you paste

The fastest fix isn’t a new subscription. It’s learning that the model doesn’t need to know who the client is to do the work.

Strip these every time: full names, street addresses tied to a name, phone numbers, email addresses, Social Security numbers, loan amounts, pre-approval letters, bank statements, and any sentence describing why the client has to move.

Replace with structure. Instead of “The Hendersons at 1420 Oak Trace need to sell because Mark got transferred to Austin and they’re pre-approved at $520K,” you write: “Seller, relocating for work, 45-day close, mid-$400s range, north-central San Antonio, wants max exposure in 30 days.” You get an identical marketing plan. You transmitted nothing.

Three tests before any paste:

Could a stranger identify this person from what’s on screen? If yes, cut more.
Would I be comfortable if this text appeared in a discovery sample? Not “will it” — “would I be comfortable.”
Does the output actually improve because of the identifying detail? Almost always no. That’s the point.

How I use this in my own business

I close 70+ transactions a year in roughly five hours a week of active management, and AI is doing real work inside that number — listing marketing suites, follow-up sequences, CMA narratives, social content. None of it touches a client’s name.

My workflow has one gate in front of it. Before anything goes into a chat window, it goes through a redaction step that converts the file into a profile: property type, price band, submarket, timeline, seller situation in three words, target buyer. Sixty seconds of work. The prompt library I use is built around those profile fields, which means the redaction isn’t a separate discipline anyone has to remember — the template simply has no field where a name would go. That’s the trick. Don’t rely on judgment at 9pm. Build the system so the unsafe version isn’t available.

The second gate is the account itself. Anyone touching client-adjacent work in my business operates in an approved workspace with training disabled, and the raw client file lives in the CRM and the transaction platform where it belongs — never in a chat thread.

When I demonstrate a full listing marketing suite built in two minutes on stage, the input is a redacted profile. Feet on the desk, coffee in hand — and nothing on that screen belongs to a real seller.

Common mistakes

Assuming the paid plan is the private plan. It isn’t. Consumer subscriptions unlock capability, not a different data agreement. The tiers excluded from training by default are the business, enterprise, education, and API contracts.

Turning off training and calling it done. The toggle is forward-looking. It doesn’t retrieve anything already used, and it doesn’t govern retention, legal holds, or subpoenas.

Uploading the document instead of typing it. Agents who carefully avoid typing a name will drag in a full pre-approval letter, a closing statement, or a CRM export without a second thought. A file upload is the same transmission with more data in it.

Letting the VA operate on a personal account. This is the single most common real exposure I see. Your policy is only as strong as the least-configured account in your business.

Using AI to touch contract language. Separate from privacy, this is a licensing issue. Drafting or modifying contract terms can constitute unauthorized practice of law under Article 13 and state statute. Keep AI on marketing, communication, and analysis — not on forms.

Frequently Asked Questions

Does ChatGPT train on my conversations?

On personal accounts — Free, Go, Plus, and Pro — yes, by default. You can turn it off under Settings → Data Controls by switching off “Improve the model for everyone.” OpenAI states it does not train on inputs or outputs from ChatGPT Business, ChatGPT Enterprise, ChatGPT Edu, or the API unless the organization explicitly opts in.

Is ChatGPT Plus private enough for client information?

No. Plus is a consumer subscription, not a business data agreement. Paying upgrades your capabilities and limits, not the legal terms governing your inputs. If you need client-adjacent data in an AI tool, move to ChatGPT Business, Enterprise, or the API, where exclusion from training is the contractual default rather than a setting you maintain.

Are deleted ChatGPT conversations really deleted?

Under current policy, deleted conversations and Temporary Chats are removed within 30 days. But in May 2025 a court order suspended that policy entirely for consumer accounts, and preserved logs from that period were later sampled for litigation. Deletion is a company policy, not a guarantee, and a court can override it without notice.

What client information should never go into an AI tool?

Names, street addresses linked to a name, phone numbers, email addresses, Social Security numbers, financial account details, loan amounts, pre-approval letters, bank statements, and the client’s reason for moving. Also avoid detail combinations that identify someone indirectly — a street, a timeline, and a life circumstance together are effectively a name in a small market.

Does using AI violate the REALTOR® Code of Ethics?

Using AI isn’t a violation. Disclosing confidential client information to a third-party platform can be, since the confidentiality obligation continues after closing. Accuracy obligations under the Code also apply to AI-generated listing content, and using AI to draft or modify contract language can raise unauthorized-practice-of-law issues under Article 13.

Should my brokerage have an AI use policy?

Yes, and NAR publishes a template brokerage AI policy covering client and consumer data, MLS content, marketing, transaction documents, CRM, and valuations. Without one, every agent makes governance decisions individually — which means your brokerage’s actual policy is whatever the least careful person on the roster decided last Tuesday.

How do I use AI on a listing without exposing the seller?

Convert the file into a redacted profile before you prompt: property type, price band, submarket, timeline, seller situation in three words, target buyer. Build that structure into your prompt templates so there’s no field where identifying data would go. The output quality is identical, because the model was never using the name.

Bring this to your team or event

Book Emily to speak at your next event:
Email: eterrell@yourcoach.com
Phone: (210) 400-9191
Web: coachemilyterrell.com

Will AI Replace Real Estate Agents by 2030? The Data

Written by Coach Emily on September 21, 2026. Posted in Uncategorized.

By Emily Terrell — Top Coach and Speaker at Tom Ferry International. Licensed since 2016. Closing 70+ deals/year while coaching agents nationwide.

No. AI will not replace real estate agents by 2030, and the labor data agrees: the U.S. Bureau of Labor Statistics projects 3% employment growth for real estate brokers and sales agents through 2034. What AI replaces is the billable justification for the admin work in your day. This guide shows which tasks are exposed.

Key Takeaways

BLS projects real estate broker and sales agent employment to grow 3% from 2024 to 2034, about as fast as the average for all occupations — the occupation isn’t contracting. bls
88% of buyers used an agent to purchase in 2025, and 91% of sellers used one to sell — the highest seller share on record. Consumer behavior is moving toward representation, not away from it. National Association of REALTORS
The exposure isn’t your job. It’s the two to three hours a day you spend producing things a model produces in ninety seconds.
Licensure, fiduciary duty, and liability are structural protections AI can’t cross without a legal change no one has proposed.
Run a replacement exposure audit on your own week before you decide whether you’re safe.

What does “AI replacing real estate agents” actually mean?

“Replacement” gets used to describe two completely different events, and conflating them is why this conversation goes nowhere.

The first is occupational replacement: software performs the licensed function, and the license disappears. That would require a statutory change eliminating agency representation in residential transactions. No state has proposed one.

The second is task replacement: specific work inside the job gets absorbed by software, and the market stops paying separately for it. That’s already happening, and it’s the only version that affects your income before 2030.

Why this matters for real estate agents

Here’s the thing nobody wants to tell you: the threat isn’t unemployment, it’s margin.

BLS counted 532,200 real estate brokers and sales agents in 2024 and projects an increase of 16,500 through 2034 (Bureau of Labor Statistics, Occupational Outlook Handbook). That’s a stable occupation, not a collapsing one. bls

Consumer behavior points the same direction. In 2025, 88% of home buyers used an agent or broker, 91% of sellers did, and for-sale-by-owner dropped to a historic low of 5% (NAR 2025 Profile of Home Buyers and Sellers). Ten years of consumer technology and the share of people going it alone hit a record low. National Association of REALTORS

Now the number that should actually worry you. According to NAR’s 2025 Technology Survey (September 2025), 17% of agents reported AI had a significantly positive impact on their business, while 46% reported no noticeable impact at all (NAR press release). Meanwhile 20% of agents use AI daily and another 22% use it weekly. nar

Read those together. Nearly half the profession is touching AI regularly and measuring nothing from it. Those agents are the ones at risk — not because a machine takes their license, but because the agent three offices over cut eight hours a week out of production and put them into appointments.

“AI is not taking your license. It’s taking the part of your week you were never actually paid for, and handing it to whoever claims it first.”
— Emily Terrell, Tom Ferry Coach

Which parts of the job are actually exposed?

Stop thinking about your job as one thing. Break it into tasks and the answer gets specific fast.

Which tasks are already gone?

Production tasks — anything where the output is text, an image, or a summary, and the input is information you already have.

Listing descriptions. Social captions. Follow-up email drafts. Market-update newsletters. Comp narratives. Open house recaps. Transaction status summaries. Property flyer copy. These are the tasks where a model gets you 80% of the way in under two minutes, and where the client never saw your labor to begin with.

You were never paid a line item for these. You were paid a commission that silently absorbed them. Which means when they take ninety seconds instead of three hours, nothing about your revenue changes. Your capacity changes.

Which tasks are structurally protected?

Four categories, and the protection isn’t sentiment — it’s law and liability.

Licensed representation. Negotiating on behalf of a principal for compensation requires a license in every state. Software doesn’t hold one, and it can’t carry the duty that comes with one.

Fiduciary judgment. Advising a seller to reject a higher offer with weak financing is a judgment call made against a specific client’s risk tolerance and timeline. That’s a duty owed to a person, enforceable against a person.

Liability absorption. When a disclosure is wrong, someone gets sued. Errors-and-omissions coverage attaches to a licensee. No consumer is going to take on that exposure to save a fee.

Physical verification and presence. The foundation crack, the neighbor conversation, the way the street feels at 6pm on a Tuesday. AI has no access to the property.

Every one of these already requires human review of AI output anyway. Fair Housing exposure in generated listing copy is real, and state advertising rules like TREC Rule 535.155 in Texas apply to AI-generated marketing exactly as they apply to anything else you publish. Review NAR’s Fair Housing resources and build the constraint into your process, not your edit pass. This is general information, not legal advice — consult your broker or attorney on specific campaigns.

What happens to the middle?

The middle is generic advice, and it’s getting commoditized in public.

“How much should I offer over list?” “Is now a good time to sell?” “What does an appraisal gap clause do?” Buyers are asking those questions in a chat window before they call anyone, and they’re getting competent answers.

That’s not a loss. That’s a buyer arriving at your first appointment already educated. The agents who lose here are the ones whose entire value proposition was explaining the process. The agents who win are the ones who show up with the local, verified, situation-specific version that no model can produce — because the data isn’t public and the judgment isn’t general.

The replacement exposure audit

Do this before your next Monday. It takes twenty minutes.

Step 1 — Log one week in three columns. Every task you touched. Column one: how long it took. Column two: could a model produce a usable first draft? Yes or no. Column three: would a client pay for this line item if it appeared on an invoice? Yes or no.

Step 2 — Add up the “yes/no” row. Tasks a model can draft that a client would never pay for separately. That’s your exposure number, in hours. Most agents I run this with land between eight and fourteen hours a week.

Step 3 — Pick one task from that row and build the workflow once. Not five. One. Write the prompt with role, context, constraints, output format, and one example of your own past work. Save it where you’ll find it again.

Step 4 — Assign the reclaimed hours before you have them. Unassigned time doesn’t become an appointment. It becomes an email. Put the hours on the calendar as prospecting or client meetings the same week you build the workflow.

The audit’s whole point is that it produces a number you can act on instead of a feeling you can worry about.

How I use this in my own business

I close 70+ transactions a year in San Antonio on roughly five hours a week of active management, and the only reason that math works is that I ran this audit on myself years before anyone was asking about 2030.

My exposure was listing marketing. I was rebuilding the same suite — description, flyer copy, social set, email announcement — from scratch for every listing. Now I build it in about two minutes and my transaction coordinator works from templates the output drops straight into. The workflow survives a week where I’m on a stage instead of at my desk, which is the entire test.

What did not change: I still walk every property. I still make the pricing call. I still sit across from the seller when the first offer comes in under ask. Nothing in my stack touches those, and nothing is close to touching them.

The agents on my roster who struggle with this aren’t struggling with the technology. They’re struggling because they built their identity around being busy, and the audit tells them roughly a third of the busy was never worth anything. That’s an uncomfortable finding. It’s also the whole opportunity. AI training that installs a workflow beats another conversation about the future of the industry every time.

Common mistakes

Treating this as a yes/no question about your job. It’s a task-level question. The person who answers it at the job level takes no action either way.
Assuming relationships protect you automatically. Relationships protect the part of the job that requires them. They don’t protect the four hours you spend on Canva.
Adopting tools instead of building workflows. 46% of agents report AI has had no noticeable impact on their business. Those agents have tools. They don’t have workflows. nar
Publishing AI output without verification. Fabricated square footage and Fair Housing drift are the two fastest ways to turn a time savings into a complaint.
Reclaiming hours and spending them on more admin. If the saved time doesn’t convert into appointments, you optimized your way into the same income with a nicer workflow.
Waiting for certainty before acting. Nobody is going to announce the date. The audit works now regardless of what 2030 looks like.

Frequently Asked Questions

Will AI replace real estate agents by 2030?

No. BLS projects 3% employment growth for real estate brokers and sales agents from 2024 to 2034, and licensed representation would require statutory change no state has proposed. AI will absorb production tasks inside the job — listing copy, follow-up drafts, market summaries — without eliminating the role that carries the license and the liability. bls

What parts of a real estate agent’s job can AI do right now?

Anything where the output is text or a summary and the input is information you already hold. Listing descriptions, social captions, follow-up sequences, comp narratives, market updates, and transaction recaps. NAR’s 2025 Technology Survey found 46% of agents already use AI-generated content, commonly for listing descriptions. Every output still needs human verification before it goes client-facing. nar

Are home buyers using AI instead of agents?

No. 88% of buyers used an agent or broker in 2025, and for-sale-by-owner sales fell to a record-low 5% of transactions. Buyers are arriving better informed because they research in AI tools first. That changes what your first appointment should cover — it hasn’t changed whether they hire someone. National Association of REALTORS

Will AI push real estate commissions down?

Possibly, but not through the mechanism most agents assume. The pressure comes from work becoming visibly cheap to produce, not from software representing clients. Agents whose value is concentrated in process explanation and content production face the most compression. Agents whose value is concentrated in pricing judgment, negotiation, and local verification face the least.

Should new agents still get licensed in 2026?

Yes, with a different plan than the one that worked in 2015. BLS projects about 46,300 openings annually over the decade. The path that fails now is building a business on tasks AI does for free. Pick a specific market or client type, build systems early, and treat AI as capacity rather than as your differentiator. bls

What should I do this week if I’m worried about AI?

Run the exposure audit above. Log one week of tasks, mark which ones a model could draft, and mark which ones a client would pay for as a line item. Build one workflow for one task in that overlap, then assign the reclaimed hours to appointments before they disappear into email.

Bring this to your team or event

Book Emily to speak at your next event:
Email: eterrell@yourcoach.com
Phone: (210) 400-9191
Web: coachemilyterrell.com

Best AI Tool for Real Estate Agents: How to Choose One

Written by Coach Emily on September 21, 2026. Posted in Uncategorized.

By Emily Terrell — Top Coach and Speaker at Tom Ferry International. 4 years coaching at Tom Ferry, 9 years prior as a client.

The best AI tool for real estate agents is the one your existing workflow already touches — ChatGPT, Claude, and Gemini all handle listing copy, follow-up, and market analysis competently. Tool choice is a smaller lever than prompt structure. This guide gives you a four-question filter for picking one, and a dated read on where the three actually differ.

Key Takeaways

The gap between agents getting results from AI and agents getting nothing is not a tool gap — 46% of agents report AI has had no noticeable impact on their business.
Pick one tool, build depth in it, and revisit the decision on a schedule instead of every time a competitor ships an update.
Four questions settle it: what job, what your stack already touches, what data you can legally put in, and what it costs to leave.
Your prompts and brand voice files are portable between all three — which means switching costs less than agents think and matters less than they fear.
What you can put into a tool is a compliance decision before it’s a productivity decision.

What is an AI tool stack for real estate agents?

An AI tool stack is the set of tools an agent uses to produce and process work — typically one general-purpose assistant for text, one video or design layer, and a CRM that receives the output. The general-purpose assistant is the piece most agents overthink, and it’s the piece that matters least in isolation.

The stack only produces value where the pieces connect. A tool that generates a follow-up email you then retype into your CRM has saved you nothing.

Why this matters for real estate agents

Adoption is no longer the problem. According to NAR’s 2025 Technology Survey (September 2025), 20% of agents use AI tools daily and another 22% use them weekly — but 46% report AI has had no noticeable impact on their business, and only 17% report a significantly positive one.

Read those together. Nearly half the agents touching AI every week are getting nothing measurable from it. That’s not a tool selection problem. That’s a workflow problem wearing a tool selection costume.

The spending makes it worse. That same survey found 24% of agents spend more than $500 a month on technology, with another 20% between $251 and $500. Meanwhile NAR’s 2026 Member Profile (June 2026) reports median business expenses climbed to $9,530 in 2025 from $8,010 the year before, against a median gross income of $59,200. Expenses are rising faster than income, and a stack of unused AI subscriptions is one of the easiest lines on that sheet to cut.

“Most agents are shopping for a tool when they should be shopping for a workflow. The tool is a $20 decision. The workflow is a 200-hour decision.”
— Emily Terrell, Tom Ferry Coach

How do you choose an AI tool for real estate?

Four questions, in order. If you answer them honestly the choice usually makes itself.

What job are you actually hiring the tool to do?

Name the task before you name the tool. “I want to use AI” is not a job. “I want listing descriptions drafted in my voice in under five minutes” is a job, and it has a testable output.

Write down the three tasks that eat the most of your week. For most agents it’s listing copy, follow-up messaging, and turning market data into something a client can read. Every major assistant handles all three. That’s precisely why the tool is not your bottleneck — your prompt is.

The right skill to build is prompt architecture, not a prompt list: role, context, constraints, output format, example. That structure transfers across every tool on this list and survives every model update. I cover the full teardown in what AI training for real estate agents must actually contain.

Which tool does your stack already touch?

The tool that lives closest to where you already work wins on adoption, almost regardless of capability.

If your brokerage runs on Microsoft 365, Copilot is already sitting in your inbox. If you live in Gmail and Google Docs, Gemini is one tab away. If your team already shares ChatGPT prompts in a group chat, that shared context is worth more than a benchmark score.

Friction beats capability. A marginally better tool that requires you to open a new tab, log in, and paste context every time will lose to a slightly worse tool that’s already open.

What data can you put into it?

This is a compliance question, and it comes before the productivity question.

Client financial details, contact records, and identifiable MLS exports do not belong in a consumer chatbot without checking two things: your brokerage’s policy and your MLS’s rules on feed usage. The workaround is simple — strip identifiers before anything touches a model. I walk through that process in how to integrate AI with MLS systems.

Generated marketing copy carries its own exposure. AI drifts toward language that signals preference based on protected class, so Fair Housing review belongs inside your prompt constraints, not in the edit pass afterward. Build the rule into the prompt and review the output against NAR’s Fair Housing resources. In Texas, AI-generated marketing is still advertising under TREC Rule 535.155 — agents in other states need the equivalent rule from their own commission.

This is general information, not legal advice. Consult your broker or attorney on specific listings and campaigns.

What does it cost you to leave?

Less than you think, which is the whole point.

Your prompts are text. Your brand voice file is text. Your listing intake template is text. All of it moves between tools in about ten minutes. The only real switching cost is the custom configuration you’ve built inside one platform — a custom GPT, a saved project, a fine-tuned instruction set — and even that is mostly re-creatable.

Knowing switching is cheap should make you commit harder, not hedge. The agent paying for three subscriptions “to keep options open” is paying a premium to avoid a decision that costs almost nothing to reverse.

Where the three tools actually differ

Current as of September 2026. Review this section quarterly — everything below it in this post is stable, and everything in it is not.

ChatGPT has the adoption lead by a wide margin. NAR’s 2025 Technology Survey found ChatGPT used by 58% of agents who use AI, followed by Gemini at 20% and Copilot at 15%. That lead compounds: more tutorials, more third-party integrations, more colleagues who can help when you’re stuck. It also generates images natively, which matters if you’re producing social graphics in the same place you write captions.

Claude is the stronger performer on long documents and sustained context — feeding it a full MLS export, a brand voice file, and six past listings in one conversation and getting consistent output back. That makes it a fit for agents building a repeatable content system rather than one-off drafts. It does not generate images.

Gemini sits inside Google Workspace, which is the entire argument for it. If your contracts are in Drive and your client correspondence is in Gmail, proximity to your actual files removes the copy-paste step that kills most AI workflows. Its free tier is generally the most generous of the three.

Copilot deserves a mention because brokerage IT departments keep choosing it. If your firm is on Microsoft 365, you likely already have it, and “already paid for” is a legitimate tiebreaker.

Here’s the thing nobody selling a comparison post wants to tell you: every one of these companies has restructured its model lineup more than once this year. Any post that ranks them by benchmark score is stale before it finishes indexing. Choose on workflow fit, which changes slowly, not on capability rankings, which change monthly.

How I use this in my own business

I close 70+ transactions a year in San Antonio in roughly five hours a week of active management, and I’m not loyal to a tool. I’m loyal to a folder of prompts.

That folder holds my listing intake structure, my brand voice file, my seller-update template, and my comp-narrative prompt. When I take a new listing, the work isn’t “which AI should I use” — it’s pulling the right prompt, dropping in the property data, and reviewing the output. Feet on the desk, coffee in hand. The whole marketing suite takes minutes because the thinking happened once, months ago, when I built the prompt.

I’ve moved that folder between tools twice. Both times it took an afternoon and nothing broke. That experience is why I push back hard when an agent tells me they’re paralyzed choosing a platform — you’re not making a permanent decision, you’re making a Tuesday decision.

The one thing I don’t move is the destination. Every output lands in a template my transaction coordinator already uses, which is why the system survives a week when I’m on a stage instead of at my desk. Automating the work around your listings is where the hours actually come back.

Common mistakes

Paying for three subscriptions and building depth in none. Three shallow tools produce less than one tool you know well. Cancel two.
Judging a tool on one bad output. A weak prompt produces weak output on every platform. Before you switch tools, rewrite the prompt with role, context, constraints, and format — then judge.
Storing prompts in chat history. If your best prompt only exists in a scrolled-back conversation, you don’t have a system. Put it in a document.
Pasting identifiable client data into a consumer tool. Strip names, contact details, and financial specifics first. Check your brokerage policy before you need it, not after.
Switching every time a competitor ships an update. The productivity cost of relearning a tool exceeds almost any capability delta between the majors. Set a review date instead — twice a year is plenty.
Treating AI output as finished. AI drafts, you decide. Every generated property fact gets verified against the MLS record before it is published.

Frequently Asked Questions

Should real estate agents use ChatGPT, Claude, or Gemini?

Use whichever one is closest to where you already work. ChatGPT has the widest agent adoption and native image generation, Claude handles long documents and consistent voice across a body of work, and Gemini sits inside Google Workspace. All three draft listing copy and follow-up competently. The difference in your results will come from your prompts, not your platform.

Which AI tool do most real estate agents use?

ChatGPT, by a wide margin. NAR’s 2025 Technology Survey found ChatGPT used by 58% of agents who use AI, Gemini by 20%, and Microsoft Copilot by 15%. That adoption lead has a practical benefit beyond capability — more tutorials exist, more colleagues can troubleshoot with you, and more third-party real estate tools integrate with it directly.

Do I need to pay for more than one AI tool?

No, and most agents who do are wasting money. NAR’s 2025 Technology Survey found 24% of agents spend over $500 a month on technology while median business expenses hit $9,530 in 2025. Pick one paid assistant, build depth in it, and add a second only when you can name the specific task the first one can’t do.

Is the free version good enough for real estate work?

For most agents starting out, yes. Free tiers handle listing description drafts, social captions, and email rewriting. You outgrow free when you need longer context — feeding a full MLS export plus your voice file into one conversation — or when usage limits start interrupting your workflow mid-task. Start free and let the friction tell you when to upgrade.

Can I put client information into ChatGPT or Claude?

Not without checking first. Your brokerage may have a policy, and your MLS may restrict how feed data is used with AI tools. The safe default is to strip client names, contact details, financial specifics, and showing instructions before anything is pasted in. Working with de-identified data removes most of the risk without removing the usefulness.

Which AI tool is best for writing listing descriptions?

All three produce comparable listing copy when given the same structured prompt. The quality difference comes from your input: property features, quality-of-life details, verified school and HOA facts, and a voice file showing how you write. Feed the same property data into two tools and compare — whichever needs less editing to sound like you is your answer.

How often should I re-evaluate which AI tool I use?

Twice a year. More often than that and you spend more time relearning interfaces than producing work. Set a calendar reminder, spend an hour testing your three most-used prompts in a competing tool, and switch only if the output is meaningfully better across all three — not just one.

Does switching AI tools mean rebuilding all my prompts?

No. Prompts are plain text and move between platforms in minutes. What doesn’t transfer is platform-specific configuration — custom GPTs, saved projects, stored instruction sets — and even those are usually re-creatable in an afternoon. Keeping your prompt library in a document rather than in chat history is what makes switching cheap.

Bring this to your team or event

Book Emily to speak at your next event:
Email: eterrell@yourcoach.com
Phone: (210) 400-9191
Web: coachemilyterrell.com

What Is a Claude Project, and How Do Agents Use One?

Written by Coach Emily on September 21, 2026. Posted in Uncategorized.

By Emily Terrell — Top Coach and Speaker at Tom Ferry International. Active San Antonio agent closing 70+ transactions a year.

A Claude Project is a saved workspace inside Claude where your instructions and reference files carry into every chat automatically. For real estate agents, that means your voice, compliance rules, and listing templates are already in the room before you type a word. This guide covers the five projects worth building first and what goes inside each.

Key Takeaways

A Claude Project stores your instructions and files once, so every new chat starts already briefed instead of starting from zero.
Projects are available on free Claude accounts, capped at five, while paid plans expand how much knowledge a project can hold and Team and Enterprise plans add sharing.
The project is only as good as the documents inside it, which makes this a documentation problem before it’s an AI problem.
Build one project per recurring job (listing copy, follow-up, social, client documents, market updates) rather than one project for everything.
Client names, financials, and contact records stay out of project knowledge, so redact first and follow your broker’s AI policy.

What is a Claude Project?

A Claude Project is a dedicated workspace in Claude with its own chat history, its own uploaded reference files, and its own standing instructions. Anything you add to the project’s knowledge base becomes context for every chat you open inside that project, and the instructions shape every response (Claude Help Center). Think of it as a briefing binder Claude reads before every conversation. You write it once, and you stop re-explaining yourself.

Why do Claude Projects matter for real estate agents?

Most agents aren’t getting anything measurable from AI yet, and the reason is almost always missing context. According to NAR’s 2025 Technology Survey (September 2025), 17% of agents reported a significantly positive impact from AI on their business, while 46% said it had no noticeable impact (NAR). The same survey found 20% of agents use AI tools daily and another 22% use them weekly.

Here’s the thing nobody wants to tell you: the tool isn’t why your output sounds generic. You open a blank chat, type “write a listing description,” and get the same beige paragraph every agent in your market is posting. Nothing about your voice, your market, your broker’s rules, or your best past work made it into the conversation. A project fixes that by making the context permanent.

Cost matters too. According to NAR’s 2026 Member Profile (June 2026), median business expenses for Realtors rose to $9,530 from $8,010 the prior year (NAR). Before you add another subscription, know that the free Claude plan includes up to five projects. That’s enough to test this entire system before you pay for anything.

“A Claude Project doesn’t make AI smarter about your business. It forces you to write down how your business works, and that one document will outlast every prompt you’ve ever saved.”
— Emily Terrell, Tom Ferry Coach

How is a Claude Project different from a Custom GPT?

The mechanics are close, and the discipline is identical. NAR’s 2025 Technology Survey ranked ChatGPT the most-used AI tool among agents at 58%, so many agents already know Custom GPTs. Both let you save instructions and upload reference files so you never start cold.

Pick the tool your team will actually open every day, and spend your energy on the documents. A voice guide, a compliance block, and five of your best listing descriptions work in either tool. The agents getting results from AI didn’t pick the right platform. They wrote down how their business works.

Which Claude Projects should a real estate agent build first?

Build one project per recurring job. A single “real estate” project stuffed with every file you own produces muddy output, because the listing-copy rules and the sphere-email tone end up fighting each other.

What goes in a listing copy project?

A listing copy project turns property specs into portal-ready descriptions in your voice, with compliance built in.

Instructions: your tone, word counts by destination (MLS remarks, portal descriptions, social), a Fair Housing constraint banning language that signals preference for any protected class, and a rule that every property fact comes from the specs you paste, never from Claude’s assumptions.
Knowledge: five of your strongest past descriptions and your listing intake form.

This is the same workflow behind automating MLS listing syndication. The project makes the saved prompt permanent.

What goes into a sphere follow-up project?

A follow-up project drafts check-ins, home anniversary notes, and past-client touches that sound like you on a good day.

Instructions: your sign-off, how casual you get by text versus email, and a rule that no first touch asks for business.
Knowledge: ten past emails and texts that actually got replies, with names and personal details stripped.

Drafts go into your CRM templates, not straight to a client.

What goes in a social content project?

A social content project produces captions, Reel scripts, and carousel copy on your content pillars.

Instructions: your pillars, your hook style, and your broker identification requirement. TREC Rule 535.155 requires broker identification on Texas advertising, and AI drafts routinely leave it out.
Knowledge: your brand guide and your ten best-performing captions.

The broker-name problem is covered in detail in AI real estate compliance: what actually risks your license.

What goes in a client documents project?

A client documents project builds CMA cover letters, buyer showing packets, and listing presentation narratives on a consistent template.

Instructions: your document structure, section order, brand colors, and fonts.
Knowledge: one finished example of each document type, with client details removed.

When a document comes out right, save it back into the project so the next one starts from the better version.

What goes in a market update project?

A market update project turns your monthly MLS stats into seller-friendly updates, newsletter sections, and social posts.

Instructions: your farm areas, the metrics you always report, and a rule that every number must come from the data you upload.
Knowledge: your stripped MLS exports and neighborhood notes, refreshed monthly.

Replace old exports instead of stacking them. Stale data in the knowledge base becomes stale claims in your content.

How do you set up a Claude Project in 10 minutes?

Go to claude.ai/projects and click New Project (Claude Help Center).
Name it for the job, such as “Listing Copy — San Antonio.” Don’t put instructions in the name or description field, because Claude can’t see either one.
Click Set project instructions and paste your standing rules: voice, format, compliance constraints, and what Claude should never do.
Upload three to five knowledge files. Pick the examples you’d be proud to publish, not everything you have.
Run one real task on an active listing, not a hypothetical, and read the output against your instructions.
Fix the instruction that caused the miss, not just the output. Then save any winning output back into the knowledge base, because chats inside a project don’t share context unless it’s saved there.

How I use this in my own business

I run my business on separate Claude Projects, one per recurring job: CMA reports, listing presentations, buyer showing packets, expired-listing letters, coming-soon marketing, and the blog you’re reading right now. My brand guide sits inside every project that produces client-facing work, so the voice and the palette are right on the first draft instead of the third.

The tool wasn’t what made it work. Writing the rules down was. When my CMA project kept producing pricing language that hedged too much for a Stone Oak seller who needed a straight answer, I didn’t rewrite the paragraph. I rewrote the instruction, and every CMA after that came out directly.

That’s the same principle behind closing 70+ transactions a year on roughly five hours of active management a week. The documents do the remembering, not me.

What should brokerage leaders know about shared Claude Projects?

On Claude’s Team and Enterprise plans, a project can be shared with specific people or the whole organization, with view-only or edit permissions (Claude Help Center). That changes the compliance picture for a brokerage.

Instead of 40 agents writing 40 versions of a listing prompt, you build one project with broker-approved instructions: Fair Housing constraints, broker identification, and your fact-verification rule. Then you share it view-only with every agent. Agents chat inside it, and only your compliance lead can edit it.

That’s the difference between an AI policy that lives in a PDF nobody reads and one built into the tool agents actually open. The live-build format for rolling this out to a room is covered in AI training for real estate agents: what it must cover.

Common mistakes

Writing instructions in the project description. Claude doesn’t read the name or description. Instructions go in the instructions field.
Building one project for everything. Rules for listing copy and rules for sphere texts conflict. One job, one project.
Uploading unredacted client files. Strip names, contact details, financials, and showing instructions before anything enters project knowledge, and check your brokerage’s AI policy first.
Never feeding the winners back in. If your best output stays buried in an old chat, every new chat starts from the same baseline. Save it to knowledge.
Leaving compliance out of the instructions. Your Fair Housing constraint and broker identification rule belong in the project, not in your memory at 10 p.m.

This is general information, not legal advice. Consult your broker or attorney on your state’s advertising and privacy rules.

Frequently Asked Questions

Are Claude Projects free?

Yes. Projects are available on free Claude accounts, which can create up to five. Paid plans (Pro, Max, Team, and Enterprise) automatically expand a project’s knowledge capacity when files approach the context limit, by up to 10 times. Sharing projects with teammates requires a Team or Enterprise plan. Five free projects is enough to build and test the core system in this guide.

What’s the difference between a Claude Project and a regular chat?

A regular chat starts from zero every time; a project starts briefed. Inside a project, your standing instructions and uploaded files apply to every conversation, so you stop pasting your voice guide and compliance rules into each new chat. Chats inside a project still don’t share context with each other unless the information is saved to the project’s knowledge base.

Is a Claude Project the same as a Custom GPT?

They solve the same problem with different tools. Both let you save instructions and reference files so the AI starts with your context. NAR’s 2025 Technology Survey ranked ChatGPT the most-used AI tool among agents at 58%, so many agents already know Custom GPTs. The documents you build, like your voice guide, compliance block, and best examples, transfer between them. Pick the tool your team will open daily.

Can I share a Claude Project with my team?

Yes, on Claude’s Team and Enterprise plans. Project creators can share with specific members by email or with the entire organization, and assign view-only or edit access. View-only members can chat inside the project and see its instructions and knowledge but can’t change them, which lets a brokerage or team leader control the compliance rules every agent works from.

Is it safe to upload client information to a Claude Project?

Treat project knowledge like any third-party tool: keep client names, contact records, financial details, and transaction specifics out. Upload templates, redacted examples, and stripped MLS data instead, and confirm your MLS permits AI use of exported data. Check your brokerage’s AI policy first. This is general information, not legal advice. Consult your broker or attorney on privacy obligations in your state.

Does Claude remember past conversations inside a project?

Partly, but don’t build your system around it. Anthropic’s documentation says context isn’t shared across chats in a project unless it’s added to the knowledge base. Claude’s memory feature does keep a separate summary for each project, but the knowledge base is the part you control. When a chat produces a winning output, save it into the project.

How many files should I put in a Claude Project?

Start with three to five, chosen for quality. A voice guide, a compliance block, and a handful of your best examples do more work than a folder dump of every document you own. Add files only when you notice Claude missing context it needs. On paid plans, projects can hold far more through retrieval, but more files don’t fix weak instructions.

Bring this to your team or event

Book Emily to speak at your next event:
Email: eterrell@yourcoach.com
Phone: (210) 400-9191
Web: coachemilyterrell.com

How Long Does AI Adoption Take? A Brokerage Timeline

Written by Coach Emily on September 21, 2026. Posted in Uncategorized.

By Emily Terrell — Top Coach and Speaker at Tom Ferry International. Licensed since 2016. Closing 70+ deals/year while coaching agents nationwide.

AI adoption at a brokerage takes about 90 days per workflow — roughly the time one repeated task needs to become routine for the agents who use it. Brokerage-wide daily use isn’t a realistic finish line, because independent contractors can’t be mandated. This timeline shows which workflows stick fastest, which stall, and how to sequence a year.

Key Takeaways

Plan on about 90 days for one AI workflow to become routine; habit research puts median formation time at 59 to 66 days, and many people take longer.
The clock starts when the workflow is attached to something agents already do every day, not on training day.
Workflows that repeat daily stick faster than workflows that happen once per listing, even when the per-listing one is easier to demo.
A brokerage can realistically install one workflow per quarter, which means four in a year at most.
Consistent means weekly use by the agents a workflow applies to, not a headcount across the whole roster.

What is an AI adoption timeline?

An AI adoption timeline is the time between introducing one AI workflow and the point where the agents apply to run it every week without being reminded. It starts at the first real repetition, not at purchase or training. It ends when skipping the workflow feels stranger than doing it.

That definition does two jobs. It makes the question measurable: one workflow, one group of agents, one behavior. And it removes the finish line most brokers are quietly holding — every agent, every tool, every day. That line doesn’t exist, and chasing it is how rollouts get declared failures at week four.

Why this matters for real estate agents and brokerage leaders

Your agents already started. Most of them just didn’t finish. According to NAR’s 2025 Technology Survey (September 2025), 20% of agents use AI tools daily, 22% weekly, 27% a few times a month, and 32% haven’t used AI in their business at all. In the same survey, 17% reported a significantly positive impact, while 46% said AI had no noticeable impact.

Read that “few times a month” group as a timeline problem, not a talent problem. Those agents tried something, got a result, and never repeated it often enough for it to become automatic. They’re stuck mid-curve.

Authority sets the ceiling. According to NAR’s 2026 Member Profile (June 2026), 86% of Realtors are independent contractors at their firms. You can’t mandate how they prospect or market. You can only make one workflow easier than the old way and keep it in front of them long enough to stick. That’s why a brokerage timeline runs slower than a team timeline, and why you plan it in quarters, not weeks.

How long does it take agents to build an AI habit?

The best evidence says two to three months for most people, with a wide spread on both sides.

In Dr. Phillippa Lally’s 2010 study with colleagues at UCL, participants took an average of 66 days to form a daily habit, with a range of 18 to 254 days, as Lally explained in a 2026 University of Surrey interview. A 2024 systematic review in Healthcare pooled 20 studies and 2,601 participants and found median formation times of 59 to 66 days, mean times of 106 to 154 days, and individual variation from 4 to 335 days (Singh et al., 2024).

“A key insight in behavioural science is that knowledge alone does not lead to behaviour change.”
— Dr. Phillippa Lally, Senior Lecturer in Psychology, University of Surrey

The gap between the median and the mean is the part brokers miss. A median near two months means a large share of your agents aren’t there yet at day 60. The slow group stretches the average to four or five months. If you plan to the median, you’ll judge the rollout at the exact moment much of the room is still forming the habit.

That’s where 90 days comes from. It’s not a magic number. It’s a planning window that covers the typical agent plus setup time, and it’s the first point where you can clearly see who’s running the workflow and who isn’t.

One honest caveat: these studies measured health behaviors like diet, flossing, and exercise, not software. The underlying mechanism transfers — repetition in a stable context builds automaticity — but nobody has published day counts for AI workflows specifically. Treat the numbers as a planning range, not a promise.

How long does AI adoption take for each workflow?

Two variables decide the speed: how often the workflow repeats, and how soon the agent sees the payoff. Everything else is secondary.

The research backs the first variable. The 2024 review found that frequency, timing, and individual choice all significantly influenced habit strength, with morning practices and self-selected habits generally showing greater strength.

Which AI workflows stick fastest?

Daily workflows with same-day output stick fastest. Think drafting follow-up messages from the CRM task list, turning showing notes into client updates, or writing the day’s social caption. They repeat every workday and the agent sees the result in minutes.

Tie the workflow to a fixed moment — first thing in the morning, before the first call — and the repetitions pile up quickly. The agent reviews every draft and sends it personally, which keeps this tier clear of automated-messaging rules.

Why does AI listing copy stick slower than it looks?

Listing copy is the easiest to demo and one of the slowest to become a habit, because it doesn’t repeat often enough. It’s also where most agents already start: NAR’s 2025 Technology Survey found 46% of agents use AI-generated content, such as listing descriptions.

The frequency math is the problem. The typical Realtor closed nine transaction sides in 2025, according to NAR’s 2026 Member Profile. That means listing descriptions happen at most nine times a year, and usually fewer. Nobody forms a habit at that frequency.

Roll it out anyway, because it builds credibility for everything that follows. Just don’t count it as your adoption workflow. For agents who list rarely, the goal is a saved prompt they can find in thirty seconds, not a habit.

Why do CRM and pipeline workflows take longer?

The payoff is delayed. An AI nurture draft or a conversation summary logged in Follow Up Boss doesn’t produce a visible win on day one — it produces an appointment in month three. Agents stop repeating behaviors they can’t see paying off.

These workflows need a scoreboard from the first week: drafts sent, replies received, summaries logged. They also depend on clean contact data, which most databases don’t have. Expect the clock to start after a cleanup, not at launch.

Why is AI texting and voice the slowest?

The clock can’t start until the consent work is done. In February 2024, the FCC confirmed that calls using AI-generated voices count as “artificial” under the Telephone Consumer Protection Act, which puts them under the same consent requirements as any artificial or prerecorded call. Automated texts carry their own consent rules.

If your agents’ contact records don’t show when and how each lead opted in, the first quarter of this workflow is an audit, not adoption. The full consent breakdown is in AI texting rules for real estate agents. This is general information, not legal advice. Confirm any outbound setup with your broker and an attorney licensed in your state.

When does the adoption clock actually start?

The clock starts on the first real repetition inside an agent’s actual workday, which is usually weeks after the training.

The training session installs the capability. The clock starts when the workflow gets attached to a trigger that already fires: a new-lead notification, a listing input sheet, the Monday sales meeting, the morning task list. Until that attachment exists, agents are relying on memory, and memory isn’t a system. Lally has made the same point about habits generally: a clear plan for when and where you’ll act beats a vague intention.

This is why session design matters. A live build on the agent’s own listing or lead, laid out in what AI training for real estate agents must cover, shrinks the gap between training day and the first real repetition to zero.

What does a 12-month brokerage AI adoption plan look like?

Run one workflow per quarter, ordered from fastest-sticking to slowest.

Quarter	Workflow	Why it goes here	What consistent looks like by quarter’s end
Q1	Daily follow-up drafting from the CRM task list, reviewed and sent by the agent	Repeats every workday with visible output the same day	Eligible agents run it weekly without a reminder
Q2	Listing marketing set from one saved prompt: description, captions, email	A credibility win that builds on the prompting skill from Q1	Every new listing goes out through the saved prompt
Q3	CRM pipeline workflow: nurture drafts and conversation summaries	Its delayed payoff needs the Q1 habit already in place	Summaries logged on every active lead
Q4	Consent audit, then inbound AI answering or automated texting	The clock can’t start until consent records exist	Documented opt-in on every contact in the automated sequence

Each quarter runs on the reinforcement structure in why most brokerage AI training doesn’t stick: one named owner, a five-minute weekly slot, a baseline, and a day-90 count. This plan is the calendar those quarters sit on.

“Judge an AI rollout at day 30 and you’ll kill it right before it works. Judge it at day 90, one workflow at a time, and you’ll know exactly what to fund next.”
— Emily Terrell, Tom Ferry Coach

Brokerages that launch four workflows in January usually end the year with none running. Brokerages that launch one per quarter usually end it with three. That’s slower on paper and faster in practice.

How is a team timeline different from a brokerage timeline?

A team can move faster because the leader controls the room. How to train your real estate team on AI in 90 days lays out a one-workflow-per-month cadence. It works on a four-person team because the leader runs the weekly meeting, sees every agent’s output, and keeps reinforcing the first workflow while the second one launches.

A brokerage doesn’t have that. Your agents are independent contractors, your meeting is optional, and your room might be 200 people across three offices. Stacking a new workflow every month at that scale spreads reinforcement too thin to hold. Plan quarterly at the brokerage level, and let the teams inside your brokerage run faster if their leaders can support it.

How I use this in my own business

The workflow I expected to stick first didn’t. I assumed listing copy would be the easy one, because it’s the most obvious demo. What actually became automatic first was my morning follow-up draft. Every weekday before my first call, the CRM task list goes into a saved prompt, I edit the drafts with my feet on the desk and coffee in hand, and they go out. Same time, same chair, every day. Within about two months I’d stopped thinking of it as using AI at all.

I see the same pattern across the agents I coach. The ones who reach consistent use fastest aren’t the most tech-comfortable. They’re the ones who picked a daily workflow and tied it to a fixed moment in their day. The ones who start with something they do twice a quarter are still “getting to it” six months later.

That’s a big part of how I close 70+ transactions a year in roughly five hours a week of active management. The tools aren’t better than yours. The repetition is.

Common mistakes

Judging the rollout at day 30. Without reinforcement a workflow decays fast, and even with reinforcement most agents need two to three months. Day 30 tells you who started, not who’ll stick.
Leading with the lowest-frequency workflow. Listing copy demos well and repeats rarely. Lead with something agents do every day.
Measuring against the whole roster. Adoption is a percentage of eligible agents. A buyer specialist who never lists isn’t failing your listing workflow.
Planning to the median. Many agents form habits slower than the median. Budget reinforcement past day 66, not up to it.
Assigning one workflow to everyone. Self-selected habits showed greater strength in the 2024 review. Offer a short menu and let agents pick.
Turning on automated outreach before the consent audit. The timeline doesn’t start until the records exist, but the legal exposure starts with the first message.

Frequently Asked Questions

How long does AI adoption take at a real estate brokerage?

It takes about 90 days per AI workflow for the agents who use it to run it routinely. Habit research puts median formation time at 59 to 66 days, and setup adds weeks. A brokerage can realistically sequence one workflow per quarter. Brokerage-wide daily use by every agent isn’t a realistic target, because independent contractors can’t be mandated.

How long does it take real estate agents to form an AI habit?

Most people need two to three months to make a new daily behavior automatic. Dr. Phillippa Lally’s 2010 UCL study found an average of 66 days, with a range of 18 to 254 days. A 2024 systematic review found medians of 59 to 66 days. Those studies measured health behaviors, not software, so treat the figures as a planning range for AI workflows.

Which AI workflow should a brokerage roll out first?

Start with a daily workflow that produces visible output the same day, such as drafting follow-up messages from the CRM task list. Daily repetition builds the habit fastest. Listing copy demos well but repeats too rarely for most agents to form a habit, since the typical Realtor closed nine transaction sides in 2025. Save it for the second quarter.

Can a brokerage get every agent using AI consistently?

No, and that shouldn’t be the goal. According to NAR’s 2026 Member Profile, 86% of Realtors are independent contractors at their firms, so a brokerage can’t mandate how they work. The realistic target is consistent weekly use of each workflow by the agents it applies to. Measure adoption as a share of eligible agents, not the whole roster.

How do you know when an AI workflow has become consistent?

A workflow is consistent when the agents it applies to run it every week without a reminder for a full month. Check it by counting usage in the last two weeks at day 90, not by surveying how agents feel. If usage drops the week reinforcement stops, the habit hasn’t formed yet and the workflow needs another cycle.

Should agents choose their own AI workflow?

Yes, from a short menu. A 2024 systematic review of habit formation found self-selected habits generally showed greater strength. Offer two or three daily workflows that fit your brokerage, let each agent pick one, and hold the rest of the rollout constant: same owner, same weekly slot, same day-90 measurement.

How is a team AI rollout timeline different from a brokerage’s?

A team can add a new AI workflow about once a month because the leader runs the meeting and reinforces the first workflow while the second launches. A brokerage of independent contractors can’t reinforce at that density across a large, optional-attendance roster. Plan one workflow per quarter at the brokerage level, and let teams inside the brokerage move faster.

Bring this to your team or event

Book Emily to speak at your next event:
Email: eterrell@yourcoach.com
Phone: (210) 400-9191
Web: coachemilyterrell.com

Real Estate AI Policy: Why Your Brokerage’s Isn’t Enough

Written by Coach Emily on September 21, 2026. Posted in Uncategorized.

By Emily Terrell — Top Coach and Speaker at Tom Ferry International. Speaker for NAHREP, eXp Con, and brokerages nationwide.

A real estate AI policy is your own one-page document governing how you use AI — and your brokerage’s version doesn’t replace it. Under TREC Rule 535.155 and the Fair Housing Act, the license holder owns the advertisement regardless of who drafted it. This guide covers the four sections you need and the tool inventory most agents skip.

Key Takeaways

Your brokerage’s AI policy protects the brokerage. Yours protects your license. They’re different documents doing different jobs.
Four sections cover almost all of your real exposure: a review gate, an input rule, a Fair Housing scan, and a rule for generated voice and media.
The failure mode isn’t a reckless agent. It’s a tool that turned on an AI feature underneath you — which is why the tool inventory matters more than the policy language.
Fair Housing doesn’t ask who wrote the listing description. It asks whether the language signals a preference based on protected class.
A policy you write in an afternoon and actually follow beats a twelve-page document you skim once.

What is a real estate AI policy?

A real estate AI policy is a written standard defining which AI tools you use, what information may go into them, what human review happens before AI-assisted output reaches a client or the public, and how AI-generated voice, images, and video get handled. At the brokerage level it’s a supervisory document. At the individual agent level it’s a workflow document — a set of gates built into how you already work.

Most agents assume it’s the brokerage’s job. That assumption is the problem.

Why this matters for real estate agents

Adoption already happened. According to NAR’s 2025 Technology Survey, drawn from a random sample of 49,233 active Realtors, 68% of agents use AI tools in their work. A February 2026 survey of 225 NAR members conducted by Realtors Property Resource put the figure at 82%, with 92% either using AI now or planning to. That same RPR survey found 63% of respondents named accuracy of outputs as their top concern.

Brokerage leadership caught up fast. According to the Delta Media Group Real Estate AI & Leadership Survey reported by HousingWire in 2026, only 1.9% of surveyed brokerage leaders had no plans to adopt AI in 2026 — down from 10.6% in 2024.

Here’s the thing nobody wants to tell you: near-universal adoption plus near-universal uncertainty about accuracy is exactly the condition where compliance problems surface. And they surface on the license holder.

TREC Rule 535.155 defines an advertisement as any communication by or on behalf of a license holder designed to attract the public to use real estate brokerage services — including email, text messages, social media, and the internet. Nothing in that rule cares whether a human or a model produced the words. Responsibility runs in both directions: your name is on the ad, and your broker is separately required to ensure your advertising complies. That’s precisely why one policy can’t do both jobs. Your brokerage’s document manages the brokerage’s supervisory duty. It doesn’t manage your workflow, your tools, or your specific listings.

The National Association of REALTORS® published guidance in March 2026 making the brokerage-level case and offering a customizable AI policy template for brokers. It’s a strong document. It’s also written for the person supervising you, not for you — and the template itself sits behind a member login.

If you’re the one running the training rather than sitting in it, the compliance segment belongs in the room, not in the appendix. I’ve broken down what AI training for real estate agents has to cover separately.

“Your broker’s AI policy is a supervision document. Yours is a workflow document. If the only AI rules in your business live in a PDF you’ve never opened, you don’t have a policy — you have a liability with a cover page.” — Emily Terrell, Tom Ferry Coach

This is general information, not legal advice. Confirm your specific obligations with your broker and an attorney licensed in your state before relying on any of it.

The four sections your own AI policy needs

What outputs require human sign-off before they leave your hands?

Write down which categories of output get reviewed and owned by a licensed human, and which are internal-only. Client-facing goes on the review list: listing descriptions, client emails, market analysis, social captions, video scripts, chatbot replies. Internal-only doesn’t: meeting notes, drafts you’re still working, research summaries you’ll rewrite anyway.

The rule is one line: AI drafts, the agent decides. Everything downstream inherits that. Name a person for each category — usually you, sometimes your transaction coordinator — so the gate has an owner instead of a hope.

What information never goes into an AI tool?

Client names paired with financial details. Loan pre-approval documents. Contract terms tied to an identifiable party. Anything you couldn’t explain to your client if they asked exactly where their data went.

Split your tools into two buckets: business-grade with documented data handling, and everything else. Consumer-tier tools may retain prompts or use them for model improvement. That’s not a reason to avoid them — it’s a reason to know which bucket you’re in before you paste. When I need AI on real client data, I strip identifiers first. The math still works; the exposure doesn’t follow it.

What language triggers a Fair Housing rewrite?

This is the section with the largest downside and the least written-down process across the industry. Build a fixed scan list and run every AI-generated marketing line through it before publishing. Catch these:

“Perfect for families,” “great for young professionals,” “ideal for retirees” — familial status and age
“Safe neighborhood,” “quiet, established area” — frequent proxies for race and national origin
“Walking distance to [church, temple, mosque]” — religion
School quality used as a neighborhood descriptor rather than a factual amenity
“Exclusive,” “integrated,” “up-and-coming” — coded community language
Any sentence describing who the home is for rather than what the home is

Better still, build the constraint into the prompt so you’re preventing the language instead of catching it. Then scan anyway. Generative models trained on decades of listing copy have absorbed decades of listing copy habits.

How do AI voice, images, and video get handled?

The FCC confirmed in a February 2024 declaratory ruling that AI technologies generating human voices fall under the TCPA’s restrictions on artificial or prerecorded voice, which means those calls require the prior express consent of the called party. Your AI-cloned voicemail drop is a robocall in the eyes of the rule. Treat it that way.

For images and video: write down whether generated or substantially altered property imagery gets a disclosure, and where that disclosure appears. Write down whether you use AI avatars of yourself, and whether they’re labeled. If you feature generated people in your marketing, note that consistently generating one demographic creates the same advertising exposure that the words do. The same broker-identification rules that govern your short-form video content apply whether the face on screen is yours or a generated stand-in.

The tool inventory most agents skip

Here’s what I’d actually bet on as the failure mode. Not an agent doing something reckless — an agent using a tool that turned on an AI feature underneath them.

Your CRM adds AI call summaries. Your marketing platform starts auto-generating captions. Your transaction management system introduces an AI document reader. None of that showed up in a decision you made. All of it now touches client data.

So the policy needs a fifth thing, and it’s a list, not a rule: every tool that touches client data, what AI features it runs, and the date you last checked. Review it quarterly. A policy that describes a stack you no longer run is a document about a business you don’t have.

For the deeper build on tool selection and MLS-data handling specifically, see how to integrate AI with MLS systems — most MLS contracts carry their own data-use restrictions that sit on top of everything here.

What about state AI law?

Texas agents ask about this constantly, and there’s a widespread misreading worth correcting.

The Texas Responsible Artificial Intelligence Governance Act (TRAIGA, HB 149) took effect January 1, 2026. It prohibits developing or deploying an AI system with intent to discriminate against a protected class, and the Texas Attorney General holds exclusive enforcement authority with a 60-day cure period. Penalties for curable violations run $10,000 to $12,000; incurable violations run $80,000 to $200,000.

But the affirmative consumer-notice duty — telling people before they interact with an AI system — landed on governmental entities and, in treatment contexts, healthcare providers. The earlier draft would have reached private companies. The enacted version narrowed that out. So no, Texas does not currently require you to disclose AI use to your clients.

That’s the law today, not the law in eighteen months. Build the disclosure capability into your policy now and you won’t be retrofitting it later.

How I use this in my own business

My own policy is one page in the same ops folder as my transaction checklist, and it gets read at the same cadence.

The section I use most is the input rule. On a Stone Oak listing last spring, the seller sent over a full financial picture — payoff figures, a second lien, and the number they needed to clear. I wanted AI help modeling the net-sheet scenarios. What went into the tool was the math with every identifier stripped: no name, no address, no loan numbers. What came back was three clean scenarios I walked her through at the kitchen table.

The review gate is the one my transaction coordinator uses. Nothing client-facing publishes without a licensed human’s name attached to it. That’s not a compliance ritual. That’s the same gate that catches the wrong square footage before a buyer’s agent does.

Common mistakes

Waiting for the brokerage. Your brokerage’s document manages their supervisory duty. It doesn’t manage your listings, your CRM, or your Instagram captions.

Writing twelve pages nobody reads. One page you follow beats a policy manual you skim once during onboarding. Length is not rigor.

Treating the Fair Housing scan as a vibe check. “I’d notice if it were discriminatory” is not a process. A fixed list is a process.

Skipping the tool inventory. The policy language is the easy part. Knowing what’s actually running is the part that expires.

Assuming the vendor absorbed the liability. Read the terms. In almost every case, the tool provides output and you provide judgment — and the license holder provides the license.

Never revisiting it. NAR guidance, state law, and your own tool stack all moved in the last twelve months. Quarterly review, on the calendar, or it doesn’t happen.

Frequently Asked Questions

Does my brokerage need an AI policy?

Yes, and so do you. Your brokerage’s policy addresses their supervisory obligation over advertising, disclosures, and client communications. NAR published broker guidance and a customizable template in March 2026. But that document governs the brokerage’s exposure, not your daily workflow, your tool stack, or the specific listings carrying your name.

What should a real estate AI policy include?

At the individual agent level: a review gate defining which outputs need human sign-off, an input rule defining what client information never enters an AI tool, a Fair Housing scan list run before any marketing publishes, rules for AI-generated voice and media, and a quarterly inventory of every tool touching client data. NAR’s broker-level template adds scope, approved tools, training, and enforcement.

Do real estate agents have to disclose AI use to clients?

No federal rule currently requires blanket disclosure that AI assisted your marketing. Specific uses do carry duties — AI-generated voice calls require prior express consent under the TCPA per the FCC’s February 2024 ruling, and altered or generated property imagery carries accuracy and misrepresentation exposure. Check your state commission and MLS rules, which vary and are changing.

Can I put client information into ChatGPT?

Not into consumer-tier tools without understanding their data handling. Many retain prompts or use them for model improvement. Use business-grade tools with documented data policies for anything involving client PII, or strip identifiers before the data goes in. The test: could you explain to your client exactly where their information went?

Does Texas law require agents to disclose AI use?

Not currently. TRAIGA took effect January 1, 2026, but its consumer-notice requirement applies to governmental entities and, in treatment contexts, healthcare providers — not to private real estate businesses. TRAIGA’s prohibition on deploying AI with intent to discriminate does apply broadly. The Texas Attorney General enforces it exclusively, with a 60-day cure period.

Who is liable if AI writes a Fair Housing violation into my listing?

You are. The Fair Housing Act addresses the advertisement, not the authorship. TREC Rule 535.155 places advertising responsibility on the license holder, and separately requires brokers to ensure sponsored agents’ advertising complies. “The tool wrote it” has no standing in either framework.

How long should my AI policy be?

One page. If it runs longer, you’ve written a document you’ll reference once. The goal is a set of gates you actually follow, not comprehensive coverage of every hypothetical.

How often should I update it?

Quarterly, tied to the tool inventory. The policy language changes slowly. Your tool stack doesn’t — features get added to platforms you already pay for, without a decision on your part.

Bring this to your team or event

Book Emily to speak at your next event:
Email: eterrell@yourcoach.com
Phone: (210) 400-9191
Web: coachemilyterrell.com

What Should Real Estate Agents Never Use AI For?

Written by Coach Emily on September 21, 2026. Posted in Uncategorized.

By Emily Terrell — Top Coach and Speaker at Tom Ferry International. Active San Antonio agent closing 70+ transactions a year.

Real estate agents should never use AI for three things: cloned-voice outbound calls, generated client testimonials, and drafting or interpreting contract language. Each one converts an AI output into a legally operative act that a license or a federal rule reserves for a human. Everything else on your desk needs a review step, not a prohibition.

Key Takeaways

Only three AI uses are categorically off-limits — everything else is a review problem, not a ban.
AI voice cloning on outbound calls falls under the TCPA, and the FCC confirmed it in February 2024.
AI-generated testimonials and reviews carry civil penalties under the FTC’s rule, effective October 2024.
Asking AI to draft an escalation clause or explain a contingency is the unauthorized practice of law in Texas, indirectly counts, and your license doesn’t cover it.
The durable fix isn’t a longer “never” list. It’s one approval gate that three categories of output have to pass through.

What does “never use AI” actually mean in real estate?

“Never use AI for X” is a task list, and task lists don’t survive the next model release. The more useful frame is that there’s almost no task AI can’t touch — but there are moments where an AI output stops being a draft and becomes a legally operative act. Those moments require a human who holds a license, or a fact that actually happened.

That distinction is the whole post. Three uses fail that test no matter how carefully you review them, because the problem isn’t quality. It’s authority. The rest of what agents worry about — listing copy, images, valuations, client data — is real exposure that gets managed with a review step, not a prohibition.

Why this matters for real estate agents

Adoption already happened. The guardrails didn’t.

RPR’s February 2026 survey of 225 NAR members found that 82% of agents currently use AI in their business and 68% use it daily or several times per week. The same survey found compliance or legal issues was the second-most-cited concern at 49%, behind accuracy of outputs at 63%, with Fair Housing concerns at 28% (RPR, February 2026).

Read those two sets of numbers together. Two-thirds of the industry is producing AI output several times a week, and half of them already know they’re not sure where the line is. That gap is not a technology problem. It’s a missing approval gate.

“The three things you can’t delegate to a model aren’t the scary ones. They’re the boring ones — a voice on a phone call, a testimonial about a closing, and a sentence in a contract. Every one of those is a legal act, and a legal act needs a person behind it.” — Emily Terrell, Tom Ferry Coach

The three uses that are actually never

Should real estate agents use AI voice cloning for outbound calls?

No. On February 8, 2024, the FCC adopted a Declaratory Ruling confirming that calls made with AI-generated voices are “artificial” under the Telephone Consumer Protection Act (FCC, February 2024). The ruling took effect immediately.

What that means in practice: an AI voice agent working your expired list, your FSBO list, or a circle-prospecting campaign is making artificial-voice calls under the TCPA. Marketing calls require prior express written consent. Not a phone number you pulled from public record. Not a lead that filled out a form eighteen months ago. Written consent, per the standard the statute sets.

This is the item on the list with the worst math, because TCPA damages accrue per call and AI voice tools are sold on volume. A product whose entire pitch is “it dials two hundred numbers while you’re at a listing appointment” is a product that scales your exposure at exactly the same rate it scales your dials.

The narrow version that works: AI voice on inbound calls you’ve been given permission to answer, or as a transcription and summary layer on calls you’re already making. The moment the AI voice initiates contact, you’re in TCPA territory.

Can agents use AI to write client testimonials or reviews?

No, and this one catches good people who weren’t trying to cheat.

The FTC’s final rule on consumer reviews and testimonials, 16 CFR Part 465, was announced August 14, 2024 and specifically targets reviews and testimonials that misrepresent they came from someone who doesn’t exist — the announcement names AI-generated fake reviews directly — or from someone who didn’t actually have the experience described. The rule allows the agency to seek civil penalties against knowing violators (FTC, August 2024).

Here’s how agents fall into it without meaning to. You had a great closing. The client said something nice in a text but won’t sit down and write a review. So you ask a model to “write a testimonial from a happy seller in Stone Oak,” drop it on a graphic, and post it. The sentiment was real. The testimonial wasn’t. That’s the violation.

The same rule reaches two other things agents do: buying reviews with incentives conditioned on writing something positive, and testimonials from insiders — officers, managers, employees, agents — that don’t disclose the connection. A “client win” post written by your own team member is covered.

What works instead: use AI to format a real testimonial. Take the client’s actual words, ask the model to tighten them for a graphic, send it back to the client, and get their approval on the version you publish. The words originate with a real person who had a real experience. That’s the entire difference.

Can AI draft or explain contract language?

No, and Texas wrote the rule in a way that makes the AI question easy to answer.

Under 22 Texas Administrative Code § 537.11(b), a license holder may not practice law, may not “directly or indirectly offer, give or attempt to give legal advice,” may not give opinions on the legal effect of contract forms, and may not draft or recommend language defining or affecting the rights, obligations, or remedies of the principals — the rule names escalation, appraisal, and contingency clauses specifically (Texas Administrative Code § 537.11).

The word doing the work is indirect. An agent who asks a model to draft an escalation clause and then forwards it to a buyer has indirectly done the thing the rule prohibits. The model isn’t licensed. Neither, for this purpose, are you.

The same applies to interpretation. “Explain to my buyer what this financing contingency means if the appraisal comes in low” is a request for a legal opinion on the effect of a contract term. It doesn’t become something else because a chatbot answered it.

The rule does leave room. Section 537.11(c)(2) preserves your ability to explain the meaning of informational items or choices in a contract form, as long as you’re not practicing law or giving legal advice. And 537.11(d)(2) permits adding or striking language when a principal specifically instructs you in writing, with the change made conspicuous. Agents outside Texas need to check the equivalent rule from their own commission — every state has one, and they’re not identical.

What works: AI as a comprehension tool for you, not an output tool for the client. Ask the model to explain a concept so you understand it better before you refer the client to their attorney. Nothing generated goes into the file or the conversation.

What people think is “never” but is actually “review”

Most of what agents put on their mental never-list belongs in a different bucket. These are real exposures managed by a review step, and I’ve covered each one in depth:

Listing copy and Fair Housing. Models will produce “perfect for young families” or “safe neighborhood” without hesitation, because that language is statistically normal in the training data. The Fair Housing Act applies to the advertisement regardless of what drafted it (NAR Fair Housing). Build the constraint into the prompt, then read every line before it goes anywhere.
Altered or generated listing images and video. Disclosure, not prohibition — and in California it’s law. The full three-tier test is in what real estate agents have to disclose when using AI.
Broker identification in AI-generated captions. The most common Texas violation happening right now, and it’s covered in what actually risks your license.
Client data in consumer AI tools. Know what the tool retains before anything sensitive goes in. Same post covers the standard.
AI valuations. Support for a CMA, never the CMA.

This is general information, not legal advice. TCPA, FTC, and state license rules vary and change. Consult your broker and an attorney licensed in your state before setting policy.

How I use this in my own business

I close 70+ transactions a year in San Antonio on roughly five hours a week of active management, and AI touches nearly everything I produce. It touches none of the three items above, and that’s a system, not restraint.

Three categories of output route through an approval gate before anything publishes: property copy, anything that makes a claim about a past client, and anything that touches the contract. Everything else goes straight out. That’s a one-page document, not a philosophy.

The testimonial rule is the one I had to build deliberately, because it’s the one that feels harmless. A past client sent me a text after closing that was better than anything I could have written. I asked her to approve a cleaned-up version for a graphic, and she did — in writing, in the same thread. That thread lives in the transaction file. Ninety seconds of work, and the testimonial is now something I can defend rather than something I hope nobody asks about.

On voice, I made a decision early and haven’t revisited it: no AI voice agent initiates a call in my business. The tools got good fast, and the pitch is genuinely compelling when you’re staring at a list of 300 expireds. The consent math doesn’t work, so the answer is no before I look at the demo.

Common mistakes

Treating the never-list as the compliance system. The list changes every quarter. The approval gate doesn’t. Agents who memorize prohibitions instead of building a gate get caught by the item that wasn’t on the list yet.

Assuming an AI voice vendor solved consent for you. Vendors sell capability. The call is placed on your behalf, to your list, for your business. Ask any vendor to show you exactly how consent is captured and documented per number, and read the answer carefully.

Publishing a testimonial with real sentiment but generated words. The FTC rule reaches the representation, not your intent. If the person didn’t say it, they didn’t say it.

Letting AI answer a client’s contract question because you were in a hurry. Speed is the reason this happens. It’s also why “the AI explained it” reads badly in a complaint.

Confusing “AI drafted it” with “AI is responsible for it.” No regulator that has weighed in has accepted that framing. You own every output the moment it leaves your hands.

Building the list and skipping the routing. Knowing the three items is worthless if nothing in your workflow forces a stop. Write down which three categories require a human sign-off and who signs.

Frequently Asked Questions

Is it illegal for real estate agents to use AI?

No. No state prohibits licensees from using AI, and no federal rule bans it in real estate. Enforcement runs through what the content says and does, not what produced it. The three genuine limits — AI voice on outbound calls, generated testimonials, and contract drafting or interpretation — are restrictions that already existed and now apply to a faster production method.

Can I use an AI voice assistant to call my leads?

Not for outbound marketing calls without prior express written consent. The FCC confirmed in February 2024 that AI-generated voices are “artificial” under the TCPA, which means AI voice calls carry the same consent requirements as any prerecorded marketing call. Inbound answering and call transcription sit on much safer ground. Confirm with your broker and counsel before deploying any voice tool.

Can AI write client testimonials for my marketing?

No. The FTC’s rule on consumer reviews and testimonials, announced August 2024, specifically covers AI-generated testimonials that misrepresent a real person’s experience, and it allows civil penalties against knowing violators. Use AI to tighten a real client’s actual words instead, then get the client’s written approval on the published version. Real experience, real person, documented consent.

Can I ask ChatGPT to write an addendum or explain a contract clause?

No. In Texas, 22 TAC 537.11 prohibits license holders from practicing law, from directly or indirectly giving legal advice, and from drafting language affecting the rights, obligations, or remedies of the parties — escalation and contingency clauses are named specifically. Routing the request through an AI tool doesn’t change the analysis. Refer the client to their attorney.

What can real estate agents safely use AI for?

Most of the work. Drafting listing copy you then verify and read for Fair Housing, first-pass follow-up emails, social captions, market summaries, content calendars, transcription, research, and prep for listing appointments. The common thread is that a licensed human reviews the output before a consumer sees it. AI drafts, the agent decides.

Do I need a written AI policy for my team?

Yes, and it should be short. Name the approved tools, name what client data never goes into them, and name the three categories of output that require a human sign-off before publishing — property copy, past-client claims, and anything touching the contract. A one-page document beats a training session nobody remembers by Thursday. Brokerage leaders should confirm the specifics with counsel.

Does using AI void my E&O coverage?

Ask your carrier directly rather than assuming either answer. Policies differ, and some are adding AI-related language. What’s consistent across carriers is that coverage responds to what you did, not what tool you used — which is another reason the review step matters more than the tool choice.

Bring this to your team or event

Book Emily to speak at your next event:
Email: eterrell@yourcoach.com
Phone: (210) 400-9191
Web: coachemilyterrell.com

AI Voiceover for Real Estate: Where the Real Risk Is

Written by Coach Emily on September 14, 2026. Posted in Uncategorized.

By Emily Terrell — Top Coach and Speaker at Tom Ferry International. Real estate’s leading voice on AI, systems, and social media.

AI voiceover is legal in real estate marketing, and no federal or Texas rule requires you to label it on a listing video. The risk isn’t the tool — it’s placement. A synthetic voice becomes a liability the moment it goes out over a phone line or speaks as someone else. Here’s the three-placement test.

Key Takeaways

AI voiceover on a listing video, Reel, or market update is a marketing decision, not a compliance event.
The same voice on an outbound call is an “artificial voice” under the TCPA, and that carries consent, identification, and opt-out obligations.
YouTube explicitly lists cloning your own voice for voiceovers and dubs as content you do not have to disclose.
Synthesizing a client testimonial is a straight violation of the FTC’s rule on consumer reviews and testimonials, which reaches AI-generated fakes by name.
Keep your real voice on the content where a seller is deciding whether to trust you. That’s a brand rule, not a legal one — and it’s the one that costs you money if you get it wrong.

What is AI voiceover in real estate marketing?

AI voiceover is synthetic speech generated from a text script and laid over marketing content — a property tour, a market update, a neighborhood explainer, a Reel. It comes in two forms that carry completely different risks: a generic synthetic narrator that belongs to no real person, and a clone of a specific person’s voice.

That distinction does most of the work. Whose voice it is, and where it plays, determines everything that follows.

Why this matters for real estate agents

Agents are producing more content on thinner margins than at any point in the last decade. According to NAR’s 2026 Member Profile (June 25, 2026), the typical agent closed nine transaction sides in 2025 with a median gross income of $59,200, while total median business expenses climbed to $9,530 from $8,010 the year before.

Read that second number again. Expenses are rising faster than income, which is exactly why an agent looks at a $29-a-month voice tool and thinks about firing the videographer.

That instinct isn’t wrong. But most agents evaluate it as a yes-or-no question about AI, and it isn’t. It’s a question about placement — and the placement that gets agents in trouble isn’t the one they’re worried about.

“Nobody gets in trouble for narrating a listing video with a synthetic voice. They get in trouble the second that voice dials a phone. The tool didn’t change. The channel did.” — Emily Terrell, Tom Ferry Coach

The three-placement test

Sort every use of AI voice into one of three buckets before you generate a single second of audio.

Is AI voiceover okay on a listing video or Reel?

Yes, and it isn’t close. A synthetic narrator over property footage, a market stat rundown, or a neighborhood explainer is ordinary marketing production. No federal statute and no Texas rule requires you to label it.

Texas agents keep hearing otherwise, and it’s wrong. The disclosure duty in the Texas AI law landed on government entities, not private licensees — I’ve covered what actually risks your license in full, and the short version is that AI-generated marketing is still just marketing under TREC Rule 535.155. Accurate, not misleading, broker name included. Same rules your written listing copy already lives under.

Platform policy is more permissive than most agents assume. YouTube’s disclosure requirements apply to realistic AI content that makes a real person appear to say something they didn’t, alters footage of a real event, or generates a realistic scene that never occurred. Its list of content creators don’t need to disclose includes production assistance, caption creation, voice and audio repair — and, named directly, cloning one’s own voice to create voiceovers or dubs.

Your own cloned voice narrating your own listing video is on YouTube’s exempt list. Not a gray area. Written down.

What happens when an AI voice goes out over a phone line?

Everything changes, and this is the placement agents don’t see coming.

On February 8, 2024, the FCC issued a Declaratory Ruling confirming that the TCPA’s restrictions on artificial or prerecorded voice encompass current AI technologies that generate human voices — meaning calls using those technologies require the prior express consent of the called party (FCC).

Translate that into an agent’s actual week. You clone your voice, script a warm check-in, and push it to 400 past clients as a ringless voicemail or an AI dialer campaign. You have just made 400 artificial-voice calls. Marketing calls to that list require prior express written consent, plus caller identification and an opt-out mechanism. The TCPA carries private-right-of-action statutory damages per call, which means the exposure scales with your list size rather than your intent.

The identical audio file was harmless on Instagram forty minutes earlier. The channel is the whole difference.

When is an AI voice never okay?

Two cases, and neither has a workaround.

Someone else’s voice. Cloning a client, a colleague, a public figure, or a competitor without written consent puts you into right-of-publicity claims and platform impersonation policy at once. YouTube separately lets an identifiable person request removal of synthetic content that looks or sounds like them.

A synthesized testimonial. This is the one agents talk themselves into, usually framed as “the review is real, I’m just having AI read it.” The FTC’s final rule on consumer reviews and testimonials, announced August 14, 2024, prohibits businesses from creating or disseminating reviews and testimonials that misrepresent that they come from someone who does not exist — AI-generated fakes named specifically — or who did not actually experience the service, and it gives the agency civil penalty authority against knowing violators (FTC). A testimonial covers verbal statements and identifying personal characteristics, which is precisely what a voice is.

If a listener would believe they’re hearing your client speak, and they aren’t, you’re inside the rule.

Do you have to disclose AI voiceover?

Mostly no, with two exceptions worth building into your workflow.

Disclose when a listener could reasonably believe they’re hearing a specific real person who didn’t say those words. Disclose when your state, MLS, or brokerage requires it — several states are moving on synthetic media, and the trigger is almost always misrepresentation rather than automation. That’s the same test I walk through in what agents have to disclose when using AI.

Blanket labeling every video “AI-assisted” isn’t compliant. It’s noise, and it trains your audience to ignore the label on the day it actually matters.

This is general information, not legal advice. Voice, advertising, and telemarketing rules vary by state and change frequently. Confirm your obligations with your broker and an attorney licensed in your state before you launch any calling campaign.

Common mistakes

Treating voice cloning and calling as one decision. They’re two. Cloning your voice is fine. Dialing with it is a regulated activity with a consent requirement attached.

Assuming an AI voice agent vendor has solved your compliance. Deploying someone else’s tool doesn’t transfer your obligation. You’re the caller.

Reading your own client’s five-star review in a synthetic voice. The words being real doesn’t fix it. The listener believes they’re hearing the client.

Using a synthetic voice on your personal-brand content. No rule against it. It just quietly removes the only thing about your marketing that can’t be copied.

Skipping the broker name because it’s “just a Reel.” TREC treats social media as advertising. An AI script won’t add it for you, and this is the violation actually happening at volume.

Labeling everything. Precision beats volume. A disclosure that appears on all 200 videos communicates nothing on the one where it counts.

Frequently Asked Questions

Is it legal to use AI voiceover in real estate marketing?

Yes. No federal law and no state real estate rule prohibits synthetic narration on marketing content, and none requires you to label it as a general matter. Your advertising obligations don’t change: the content must be accurate, not misleading, and carry your broker identification under rules like TREC 535.155 in Texas. The tool is unregulated; the advertisement is not.

Do I have to disclose that a video uses an AI voice?

Usually not. YouTube requires disclosure for realistic AI content that makes a real person appear to say something they didn’t, and specifically exempts cloning your own voice for voiceovers and dubs. Disclose when a listener could believe they’re hearing a specific real person who didn’t speak those words, and check your state, MLS, and brokerage rules separately.

Can I use an AI version of my voice for client follow-up calls?

Not without consent. The FCC confirmed in February 2024 that AI-generated voices fall under the TCPA’s artificial-or-prerecorded-voice restrictions, which require prior express consent of the called party. Marketing calls require prior express written consent, plus identification and an opt-out. This applies to ringless voicemail and AI dialers, and statutory damages accrue per call.

Is AI voiceover allowed on Instagram and TikTok?

Both platforms require labeling for realistic synthetic content depicting real people, events, or voices, and both apply automatic labels in some cases. A generic synthetic narrator over property footage is not what those policies target. Platform rules also change faster than statutes, so verify current policy before you build a workflow that depends on one interpretation.

Can I use AI to read a real client testimonial?

No. The FTC’s rule on consumer reviews and testimonials prohibits creating or disseminating testimonials that misrepresent who is speaking, and it names AI-generated fakes directly. A testimonial covers verbal statements and identifying personal characteristics. If a listener would believe your client is speaking and your client isn’t, the words being genuine doesn’t cure it. Use text on screen instead.

Does AI voiceover hurt engagement on real estate content?

It depends entirely on the content’s job. Informational content — stats, process explainers, market recaps — performs on clarity, and viewers rarely care who narrates. Personal-brand content performs on recognition, and a synthetic voice removes the recognition. The practical rule is to keep your voice on anything where you are the reason someone is watching.

What’s the safest way to start using AI voice in my business?

Start with one utility format and one hard boundary. Pick a repeatable, non-personal video type — a monthly market update works well — script it, generate it, and batch a month at once. Write the boundary down before you start: nothing generated goes out over a phone line, and nothing generated speaks as a person. Everything else is a production choice.

Bring this to your team or event

Book Emily to speak at your next event:
Email: eterrell@yourcoach.com
Phone: (210) 400-9191
Web: coachemilyterrell.com

AI Real Estate Video: What You Must Disclose in 2026

Written by Coach Emily on September 14, 2026. Posted in Uncategorized.

By Emily Terrell — Top Coach and Speaker at Tom Ferry International. Licensed since 2016. Closing 70+ deals/year while coaching agents nationwide.

AI video disclosure for real estate means labeling any video where AI altered how the property looks or who appears to be speaking. Basic edits — captions, color, trimming — need no disclosure. Virtual staging inside a walkthrough, replaced skies, and AI avatars do. This guide covers the four rules that apply and the disclosure checklist to run before you post.

Key Takeaways

The line isn’t AI versus no AI. It’s whether the edit changed the property or changed who’s speaking.
Most agents are citing the wrong rule. The provision that governs altered listing imagery is NAR Standard of Practice 12-10, not 12-1, 12-5, or 12-8.
Disclosure rules were written about photos. Video sits in a gap where the general duty still applies and the specific guidance mostly doesn’t.
An AI avatar delivering your listing copy triggers Article 12’s separate requirement that recipients know they’re hearing from a real estate professional.
Build the disclosure into your export template once. A rule you have to remember on a busy listing day is a rule you’ll break.

What is AI video disclosure in real estate?

AI video disclosure is the practice of telling viewers, on or beside the video itself, that artificial intelligence altered what they’re seeing. It applies when AI changes a material element of the property — furniture that isn’t there, a sky that wasn’t that color, a fence that doesn’t exist — or when the person on screen is a generated avatar rather than a recorded human.

It does not apply to production quality. Color correction, exposure, stabilization, auto-captioning, filler-word removal, and trimming are edits to the recording, not to the property. Those need no label.

The test is one sentence: did the edit change what a buyer would find when they walk in the door?

Why this matters for real estate agents

Here’s the thing nobody wants to tell you: the industry adopted AI video faster than it read the advertising rules, and the exposure is landing on the license, not the software.

According to an RPR survey of 225 NAR members (February 2026), 82% of agents now use AI in their business, while 49% cite compliance as a primary concern and 28% specifically name Fair Housing. Adoption outran governance. That gap is where complaints get filed.

The reach is real, too. According to NAR’s 2026 Member Profile, agents professionally use Facebook (76%), Instagram (57%), LinkedIn (55%), YouTube (31%), and TikTok (16%). Every one of those is a place where a listing video is advertising, and advertising is regulated.

“Agents keep asking me which AI video tool is safe. That’s the wrong question. No tool is safe or unsafe — your export process is. If the disclosure isn’t baked into the template, you will forget it on the listing that matters most.” — Emily Terrell, Tom Ferry Coach

This is general information, not legal advice. Rules vary by state and by MLS. Confirm specifics with your broker, your MLS, and an attorney licensed in your state.

The four rules that actually apply to AI video

Which NAR rule governs altered listing imagery?

Standard of Practice 12-10. Not 12-1, not 12-5, not 12-8 — and you’ll find all three miscited across the industry blogs currently ranking for this topic.

Article 12 of the NAR Code of Ethics requires REALTORS to present a true picture in advertising and marketing. Standard of Practice 12-10 extends that duty explicitly to Internet content and images, and its fifth clause prohibits “otherwise misleading consumers, including use of misleading images.” Adopted in 2007, amended in 2018 — years before generative video, and broad enough to cover it.

For the record: 12-1 governs claims that your services are free. 12-5 governs disclosing your firm’s name in advertising. 12-8 governs keeping website information current. Useful rules. Different rules.

Does Article 12 apply to an AI avatar of me?

Yes, and through a second clause most agents have never read.

Article 12 doesn’t stop at “present a true picture.” It also requires that a REALTOR’s status as a real estate professional be readily apparent in their marketing, and that recipients of real estate communications are notified that communications come from a real estate professional. A generated avatar delivering your market update is a real estate communication. The disclosure obligation travels with it.

Practically: a HeyGen-style avatar of yourself, used for repeatable content like market recaps or FAQ answers, is defensible with a label. An avatar of a person who doesn’t exist, presented as an agent, is not.

What does state advertising law require?

The same thing it always required, applied to a new format. In Texas, TREC Rule 535.155 defines advertising broadly enough to include electronic media, social media, and the Internet, and prohibits advertising that is misleading or likely to deceive the public. Your broker’s name has to appear, and on social platforms that disclosure can live on a directly linked profile page.

Nothing in that rule mentions AI. Nothing needs to. A misleading video is a misleading advertisement regardless of what produced it, and enforcement runs up through the broker.

Agents outside Texas: pull your own commission’s advertising rule and read the definition of “advertisement.” Most are written just as broadly.

What did California change, and why should agents in other states care?

California AB 723 took effect January 1, 2026, adding Section 10140.8 to the Business and Professions Code. It requires brokers, salespersons, and anyone acting on their behalf to disclose when a property image in an advertisement has been digitally altered, and to provide a link, URL, or QR code back to the original unaltered image. It carves out basic edits like lighting and cropping. Willful violations can be charged as a misdemeanor under California’s licensing law.

Inman’s coverage and Pillsbury’s summary of California’s 2026 AI laws both make the same point: this moved altered imagery from an MLS courtesy to a licensing matter with criminal exposure at the far end.

Now the part that matters for this post. AB 723 is written about images. So are most MLS altered-photo policies. Video walkthroughs, AI-generated B-roll, and avatar presenters sit in a gap where the specific statute may not reach but Article 12’s general duty absolutely does. Treating video to the photo standard is the conservative play, and it’s the one I’d make.

Where does Fair Housing enter a video?

Through the people AI puts in it.

Standard of Practice 10-3 prohibits printing, displaying, or circulating any advertisement about selling or renting a property that indicates a preference, limitation, or discrimination based on a protected class. Standard of Practice 10-1 separately bars volunteering information about the racial, religious, or ethnic composition of a neighborhood.

AI-generated B-roll of a neighborhood populates itself with people. If you prompt for “families enjoying the neighborhood” and the model returns a uniform set of faces, you just circulated a visual signal about who belongs there. Nobody wrote a discriminatory word. The exposure is the same.

This is why I tell agents to film their own neighborhood footage or license real stock, and to keep generated humans out of listing marketing entirely.

How I use this in my own business

I run 70+ transactions a year in about five hours a week of active management, and every one of my listing videos goes through the same three-line check before it exports.

Here’s the actual how. My CapCut export template has a permanent lower-third slot that carries my name, my broker’s name, and my license number — so the TREC disclosure ships automatically instead of depending on my memory at 9pm. The second slot is conditional: if any frame contains virtual staging or an altered exterior, a “Digitally altered — original photos available on request” card goes on the front of the video and the same line goes in the caption. Third, the original unedited footage stays in a dated folder in Drive, so I can produce it in under a minute.

On a Stone Oak listing this spring, the seller asked me to have the dead spring lawn “fixed” in the walkthrough video. I told them yes, and then told them it would carry a disclosure card. They were fine with it. The buyers were fine with it. Nobody has ever walked away from a listing because I told them the truth about a lawn.

Common mistakes

Disclosing the tool instead of the alteration. “Made with AI” tells a buyer nothing. “Virtually staged — this room is unfurnished” tells them what they need.
Putting the disclosure only in the MLS. The video travels to Instagram, YouTube, and your website. The disclosure has to travel with it, embedded in the frame.
Using generated people in neighborhood B-roll. This is the single highest-risk AI video habit in the industry right now, and almost nobody is talking about it.
Deleting the original footage. Every disclosure regime that has landed so far requires you to produce the unaltered version. If you can’t, the disclosure is worthless.
Treating video as exempt because the statute says “images.” The specific rule may not name video yet. The general duty to present a true picture already covers it.
Making it a memory task. Build it into the export template and the listing checklist. A rule that lives in your head is a rule you’ll skip on the busy week.

Frequently Asked Questions

Do I have to disclose AI-generated real estate videos?

You have to disclose the AI that changed the property or the presenter. Virtual staging inside a walkthrough, replaced skies, removed power lines, altered landscaping, and avatar presenters all require a label. Captions, color correction, trimming, and filler-word removal do not, because they change the recording quality rather than what a buyer will actually find at the property.

What NAR rule covers AI-altered listing images?

Standard of Practice 12-10, under Article 12. Its fifth clause prohibits misleading consumers, including through the use of misleading images, and it applies expressly to Internet content. Much of the content currently ranking on this topic misattributes the rule to 12-1, 12-5, or 12-8, which govern free-service claims, firm-name disclosure, and website currency respectively.

Does California’s AB 723 apply to video walkthroughs?

AB 723 is written about images in advertisements and promotional materials, and it hasn’t been clearly tested against video. That ambiguity is not protection. NAR Article 12 and your state’s advertising rule already cover misleading video, so the safe practice is to apply the same disclosure and original-file standard to walkthroughs that AB 723 requires for photos.

Can I use an AI avatar of myself in listing marketing?

For repeatable, non-property content like market updates and FAQ answers, yes, with a visible label identifying it as an AI-generated presentation. Article 12 separately requires that recipients know they’re receiving a communication from a real estate professional, so your name, brokerage, and license information still have to appear. Keep your real face on trust-building content.

What happens if I skip the disclosure?

Depending on where you practice, exposure ranges from an MLS violation and listing removal, to a Code of Ethics complaint under Article 12, to state licensing discipline for misleading advertising, to civil liability if a buyer claims they were misled. In California, willful violations of AB 723 can be charged as a misdemeanor under the licensing law.

Does AI-generated B-roll create Fair Housing risk?

Yes. Standard of Practice 10-3 prohibits advertising that indicates a preference or limitation based on a protected class, and generated footage of “people enjoying the neighborhood” makes that statement visually. No discriminatory language is required for the signal to land. Film real footage or license real stock, and keep generated humans out of listing marketing.

How do I build disclosure into my workflow?

Make it structural, not situational. Add a permanent lower-third to your export template carrying your name, broker name, and license number. Add a conditional disclosure card triggered by any property alteration. Save original unedited footage in a dated folder so you can produce it on request. Then add one line to your listing checklist: disclosure applied, original archived.

Does this apply to agents outside California and Texas?

Yes. Every state real estate commission has an advertising rule prohibiting misleading advertising, and NAR’s Code of Ethics binds every REALTOR regardless of state. California and Texas are examples, not the boundary. Pull your own commission’s advertising rule and your MLS’s altered-image policy, and apply the stricter of the two.

Bring this to your team or event

Book Emily to speak at your next event:
Email: eterrell@yourcoach.com
Phone: (210) 400-9191
Web: coachemilyterrell.com