Free class · Clients and follow-up
Your clients are already asking AI about price, offers, and inspections. This class shows you how to get there first, check an AI answer with your client instead of against them, and hand every new client a one-page guide that routes their AI use back through you.
Part 1
By the time a client quotes a chatbot, a wrong answer has already set. A seller who has read a number from an app will defend it, because it feels like their discovery. Arguing just makes you the obstacle, and people route around obstacles. The real move is to never have that argument: be the first AI conversation your client has.
You don't have to out-nerd ChatGPT. You bring two things it will never have. Deep local knowledge: your market, your contracts, your neighborhoods. And the ability to build something reliable with the tool and put it in your client's hands. A nervous client relaxes the moment they watch you build something that works.
You, your client, and the AI sit on the same side of the table, aimed at one thing: the right answer and the right home. Four moves get you there. The first two build the AI for your client so the bad answer rarely happens. The last two hold the line when an answer is high-stakes.
Be the first AI conversation your client has. Bring it up yourself, find out how they already use it, and hand them prompts you trust before they go find their own.
Never accept the first answer. Make the AI argue against its own work, with your client beside you, so the two of you are testing the tool instead of each other.
Nothing high-stakes leaves your hands until it clears a checkpoint: every claim listed, every source named and local, every number checked.
AI drafts the structure and the first pass. You own the part where the money and the truth live: judgment, local facts, and the relationship.
Part 2
The cheapest moment to shape how a client uses AI is before they've used it badly. Do these three things at the first meeting, whenever the moment fits.
“Most of my clients use ChatGPT for this. Let me show you how to use it well, and where it'll burn you.” Now you're framing the tool, not fighting it later.
“How are you already using AI for your move?” Then listen. A power user, a skeptic, and a nervous first-timer each need a different touch.
“Here are three I use. Text me what comes back.” Their AI use now routes back through you.
Makes you duller
Makes you sharper
Run the stress-test with your client sitting next to you. You're not telling them they're wrong. You're saying “let's see if we can break this together.” A client who would dig in to defend a number will happily help you poke holes in it when the AI is the one being questioned.
And when a client does bring you an answer you can't check on the spot, you don't need one. “Great, can you send that to me? I'll look it over and get back to you.” Then ask the question most people never ask: where did the AI get that? Links and citations either hold up or they don't.
Part 3
Read every AI output like it's trying to get away with something. Does it answer what was asked? Is every fact, number, date, and address real and local? A claim with no named source, no checkable number, and no local grounding isn't ready to leave the room. Three questions force an answer to prove itself or fold.
No source, or a national average dressed as local, means the claim is unverified. National is almost always wrong for your street.
This makes the model expose its own assumptions. If you can check the thing it names, you just found the weak point.
The confident sentence hides the error. The comp, the figure, or the clause underneath reveals it.
The validation gate
ROLE You are my validation gate. No AI output about a real estate deal leaves my hands until it clears you. You are strict and you show your work. CONTEXT - My market: [CITY / NEIGHBORHOODS] - Acceptable sources: [MLS / county records / named local data]. A national average is NOT acceptable as a local claim. INPUT Here is an AI output I am about to use with a client: [PASTE THE OUTPUT] TASK Run it through the gate and return a verdict table: 1. CLAIMS. List every factual claim, number, date, and address it makes. 2. SOURCE. For each, state whether a real, checkable, LOCAL source is named. If not, mark it UNVERIFIED. 3. INVENTED. Flag anything that looks made up, or a national stat dressed as local. 4. VERDICT. SHIP, FIX, or KILL. If FIX, list exactly what I must verify or add first. OUTPUT The table, then one line: "Ships only after you personally confirm: [...]." CONSTRAINTS Assume the AI invents confident numbers. Default any unsourced number to UNVERIFIED. Never pass a value, legal, or inspection claim without a named local source.
Paste in a price opinion, a listing description, or an offer email. You get back every claim, whether each has a real local source, what looks invented, and a verdict with the exact list of what you must confirm yourself first.
AI drafts the structure and the busywork fast. Use it. The part it can't do is judgment, local truth, and the relationship, and that part is the reason your client needs you and not just the app. At these four, slow down, pull the primary source, and let a professional make the call.
A model guessing from national data can badly mis-price a deal. You bring the comps that matter.
Disclosures, contingencies, clauses. Wrong here is liability. Route it to the contract and your broker.
Whether to waive, what to flag. That's safety and money. A person decides.
A confident AI phrasing can create real legal exposure. You write or approve every word.
The class ends by building one page that pulls all four moves together, written in your voice, that you give every new client at the first meeting. It reads like a gift, not a leash.
Free with a student account
Create a free student portal account to get the workbook, prompts, templates, and downloads for every class.
Create my free account →Already a student? The same link signs you in.
Don't argue with the number. Thank them, and ask them to send you the exact question they asked and the answer they got, links included. Then look at it together. Where did the number come from? Is it local? Is it current? Walk through your comps and show what holds up and what doesn't. You're after a shared answer, not a win.
No. Telling a client not to trust AI makes you sound like the travel agent who told people not to trust the internet. Show them how to use it well and where it burns people. The agent who fights the client's AI becomes the obstacle. The agent who teaches them to use it becomes the first call.
It means asking the AI to attack its own answer instead of agreeing with you. Ask for the strongest objections a sharp local expert would raise, the local facts that would prove it wrong, and the one assumption nobody checked. Run it with your client in the room and it stops feeling like a correction.
No. Everyone has the same access to the same tools. Your edge is knowing your market, your contracts, and your streets, and being able to build something reliable with AI and hand it over. You never have to answer on the spot. "Send it to me, I'll look at it and get back to you" is a complete answer.
Four things: what a home is worth, legal and contract terms, inspection decisions, and anything that touches Fair Housing. A confident wrong answer in any of them costs real money or creates real liability. Slow down, pull the primary source, and let a professional make the call: you, the contract and your broker, or the inspector.
Clients and follow-up
Turn a Google Sheet into a relationship memory, let AI organize and draft, and reach out only with a real, verified reason.
Clients and follow-up
Find the one useful next action for a contact, draft only when permission and facts support it, and record what really happened.
Clients and follow-up
Reset client expectations with one boundary message, 8 ready replies in your voice, and a Monday update that gets ahead of check-in texts.
Library
The whole library of recorded classes, by topic.
Research sources
Why fluent, confident output is not evidence, and why an answer has to be checked against a source.
Original research finding that citations in generative search often fail to fully support the statements they are attached to.
NAR's current overview of AI use in real estate and policy priorities around privacy, Fair Housing, and copyrighted listing content.
Brokerage governance, human oversight, and defined permitted uses for AI-assisted real estate work.
The federal rights and obligations every AI-assisted client communication still has to respect.
Model-neutral guidance for generative-AI risk, testing, documentation, transparency, and content provenance.
Product access, limits, local listing rules, platform policies, and laws can change. Re-check the linked authority before important or public work.
Keep building
AI Acceleration: Real Estate is a private, self-directed curriculum for practical AI work across content, communication, lead follow-up, CRM, client research, operations, and relationship management. Build it. Review it. Keep what works.
Explore the Skool community ↗Follow Ryan + AI Acceleration
No ranking, revenue, closing, or time-saving promise. The classes teach the work and the review systems around it.