🤖 The "AI Sold My House" Objection Script for Real Estate Agents

GROW WITH LIBERTAS & EXP REALTY

🎧 Listen on Apple Podcasts and Spotify!

The story is now famous. Stuart Thompson in New York and Robert Lavine in Florida both used ChatGPT to sell their own homes — for tens of thousands more than the agents allegedly quoted them. Both stories went viral. Sellers are pulling up these articles at your listing appointments and using them as leverage.

Most agents don't have a clean answer. Here's the breakdown that handles the objection, wins the listing, and turns the AI story into the reason the seller picks you instead of going it alone.

The framing of these stories matters. Reported one way, they sound like proof that AI is replacing real estate agents. Reported honestly, they tell a completely different story — one where AI assisted FSBOs who already had unusual professional advantages, and one where the actual sale outcomes were dramatically worse than the headlines implied.

Here's everything you need to know to be the calm, factual voice at the kitchen table.

Start with the obvious context the articles tend to bury below the headline.

Stuart Thompson is a technology reporter who writes about AI for a living. His job is producing content about AI tools. Using ChatGPT to sell his own house was both a personal transaction and a content opportunity for his career. He had professional incentive to make the experiment look successful regardless of the actual outcome.

Robert Lavine is the CEO of an AI consulting company. His entire business is built on positioning himself as an authority on AI implementation. His FSBO experiment was effectively a marketing case study for his consulting practice.

These were not the average sellers your next listing appointment is going to be. These were professional AI power users with extensive technical sophistication, deep familiarity with the tools, and significant career incentives to publicize the experiment as a win.

When your seller pulls these articles up and says "these guys did it without an agent," the honest reframe is: yes, two professional AI specialists with strong financial incentives to make the story look good were able to navigate a FSBO transaction with AI assistance. That's not the same as the average homeowner being able to replicate it.

The viral coverage almost universally left out a series of inconvenient details that change the entire picture.

Both sellers hired real estate attorneys. They paid attorney fees — sometimes thousands of dollars — to handle contract review, disclosure compliance, and legal protections. The "AI sold my house" story was actually "AI plus a paid attorney sold my house."

Both sellers paid for professional photography. They hired vendors for marketing materials, listing photos, and in some cases videography. The "AI sold my house" story was actually "AI plus paid photographers plus paid vendors sold my house."

Both sellers used the MLS. They paid flat-fee MLS services to get their listings in the same database every real agent uses. The "AI sold my house" story was actually "AI plus an MLS subscription sold my house."

Both sellers interacted with buyer's agents. They negotiated, drafted offers, and communicated with licensed professionals representing the buyers. The "AI sold my house" story was actually "AI plus the buyer's agent's expertise sold my house."

Both sellers made notable mistakes. Stuart Thompson admitted that Google's Gemini gave him incorrect advice about how to compensate buyer agents — a mistake that could have cost him the deal or created legal exposure. Robert Lavine reportedly proposed negotiation terms based on ChatGPT advice that the buyer's agent rejected because they didn't fit Florida real estate practice. AI made specific, verifiable mistakes in both transactions.

In neither case was this a "pure AI" sale. It was a FSBO supplemented with the same vendors and professionals every FSBO uses, plus an AI tool layered on top to help with research and copywriting.

Here's the part the headlines really buried.

Independent analysis of one of the transactions suggested the seller netted approximately 14% less than the comparable market value suggested by professional CMAs and recent neighborhood sales. The "sold for more than the agent quoted" framing measured the sale against an initial verbal pricing opinion — not against what the home would have actually achieved on the open market with full representation, marketing, and negotiation.

Translated to dollars on a $750,000 home — that's potentially $100,000+ left on the table. The seller saved roughly $20,000 in commission and lost five times that amount in undermarket pricing. The headline was AI helped me sell my home. The reality was AI helped me net less money than a competent agent would have.

This is the data point that wins the conversation. Frame it neutrally and let the seller do the math.

It's important to be honest about what AI is changing in this category, because pretending it's not happening is the fastest way to lose credibility with a sophisticated seller.

Two things are genuinely different now:

One — AI lowered the friction of FSBO research. Things that used to require hours of Googling, calling around, and reading state-specific guides can now be done in 15 minutes of ChatGPT prompting. Contract templates, disclosure requirements, basic pricing analysis, marketing copy — all dramatically easier than they were five years ago. Sellers who were already inclined toward FSBO now have a faster on-ramp.

Two — AI is being used by sellers to evaluate you. Before your listing appointment, most sellers are now running comps through ChatGPT, asking AI to generate questions to challenge your CMA, and using AI to research your background and reviews. You're walking into a more prepared, more informed, more skeptical conversation than you were two years ago.

Both of these are real. Pretending otherwise puts you behind the curve. The right response is to acknowledge what changed, then redirect to what hasn't.

Here's the structural truth that wins these listings.

AI can pull comps. It cannot walk the property. It cannot read whether the seller is bluffing about their walk-away price. It cannot recognize when a buyer is over-leveraged and likely to fall apart in financing. It cannot sit at the kitchen table when the appraisal comes in low and negotiate three parties back into the deal. It cannot read the room when an inspection finding threatens the entire transaction. It cannot know that the new construction down the street is offering builder financing that's pulling buyers out of the resale market. It cannot tell the seller's spouse who's on the fence about moving that everything is going to be okay.

The real work of real estate isn't the analytical work. It's the judgment-and-emotion work that lives between the data points. The negotiating call where you read a hesitation in the buyer's agent's voice and pivot the conversation. The honest conversation with the seller about why their aspirational price isn't going to work. The relationship-building with three other listing agents in the neighborhood so when their buyers don't write on their listings, your phone rings.

None of that is in any AI tool's training data. None of it ever will be — not because the technology isn't impressive, but because that work fundamentally happens in real-time human-to-human interaction with stakes that AI doesn't have to live with.

The buyer or seller, when push comes to shove, doesn't want an AI that's never been accountable for an outcome. They want a human who is.

Every listing appointment in 2026 is being prepared for by your seller using AI in advance. Assume the following list of questions has already been asked:

You're walking into a conversation where the seller has read seven articles, run three comparative analyses, and prepared a list of objections. The right move is not to be defensive. The right move is to be prepared for exactly that conversation.

This is the single most important fact to internalize and be able to articulate calmly to a seller.

AI pulls historical data. Real estate is a forward-looking transaction. The gap between those two facts is where 80% of the seller's AI analysis falls apart.

A comp that closed 60 days ago closed under different market conditions. There may have been 20 competing listings that month. There may have been a competing builder pulling buyers away. Interest rates may have been higher. A specific mortgage product may not have existed yet. A nearby school district may have just changed boundaries.

AI doesn't know any of that. It sees a closed price and reports it. The number is technically accurate. The conclusion drawn from it is often wildly wrong.

Your competitive advantage is that you walked through three of those competing listings last week. You know what condition the comp was in. You know which one had the deferred maintenance. You know which one priced aggressively because of a divorce. You know which one took 90 days and which one took 12. You know which buyers are currently active in the price range and what they're looking for.

That information cannot be pulled from any AI tool. It can only be developed by an agent who is genuinely active in the local market every single week.

When the seller pulls up the Stuart Thompson or Robert Lavine article and says "these people sold their homes using AI — why do I need an agent?" — here's the response:

That response does five things at once. It validates the seller's research without conceding that the conclusion was correct. It introduces the missing context calmly and factually. It positions the AI tool as a real but limited input. It separates the analytical work AI can do from the judgment work only you can do. And it ends with the seller agreeing to a fair-sounding next step.

Here's the deeper truth that should make you optimistic.

The AI-FSBO trend is going to create a small increase in the number of homeowners attempting to sell without an agent. Most will fail. Most will return to the market needing professional representation after a frustrating few weeks of trying to manage showings, navigate offers, and respond to inspection issues without expertise.

Failed AI-FSBOs are about to be one of the best new lead sources in real estate. Track expired FSBO listings the same way you track regular expireds. Have a script ready for the I tried this myself and it didn't work conversation. These sellers are humbled, motivated, realistic about commission, and ready to hire competence.

The AI-FSBO wave isn't a threat to skilled listing agents. It's a temporary detour for sellers who will end up in your pipeline a few months later, only now they're significantly more committed to the listing relationship.

If there's a category of agent who should be paying attention to the Stuart Thompson story, it's the average agent who has been getting by on relationship-only listings, surface-level CMAs, and minimal market knowledge.

Sellers in 2026 can now generate a serviceable comparative analysis in 15 minutes. If the entire value you brought to a listing appointment was the printed CMA, you've been commoditized. If your listing presentation was 70% slideshow with stock photos of houses you've never seen, you're in trouble. If the depth of your local market knowledge is identical to what ChatGPT can produce with five prompts — the seller now knows that.

The agents who win the next decade are the ones who:

The agents who lose are the ones whose value proposition was always shallow and is now exposed. AI didn't make agents obsolete. AI made the average agent obsolete. The skilled agent is more valuable than ever.

There's an oddly hopeful truth buried in all of this. AI is going to free skilled agents from the work they never wanted to be doing in the first place — and force them to spend more time on the work they actually got into this business for.

The mundane CMA generation, the property description writing, the marketing copy drafting, the transaction coordination paperwork, the scheduling, the recurring social posts, the answering of basic buyer questions, the data lookups — all of that is going to be handled by AI tools that will only improve from here. What's left for the agent is the relationship work. The advisory work. The negotiation work. The judgment work. The work that human beings actually want to do, that creates lasting client relationships, and that produces the thank you notes that no transactional service can ever inspire.

The agents who fight AI lose. The agents who embrace AI as the engine that frees them to be more present, more human, more available, more skilled at the truly important parts of their job — they're about to have the best years of their careers.

The viral AI-FSBO stories are interesting. They are not what they appear to be on the surface. The sellers were professional AI specialists with strong incentives to publicize the experiment. They paid for attorneys, photographers, MLS access, and buyer-agent representation. They made specific mistakes that cost them in negotiation. And the actual sale prices, when measured honestly, were below market value by margins that dwarfed the commission they avoided.

Your next seller is going to bring up these stories. Be ready. Acknowledge the parts that are true. Bring the missing context. Reframe the conversation around the work AI cannot do. Position yourself as the human accountable for the outcome.

The listing is yours to win — if you don't make the mistake of arguing with AI on data instead of competing with AI on judgment.

Ready to stop guessing and start producing?

🎯 Start Premier Coaching (free trial): premiercoaching.com
📲 Elite Coaching — text Tim directly: 512-758-0206

When the seller pulls up the AI sold my house article at your next listing appointment — are you going to argue with the data or use it as the bridge to the conversation about what only you can do?

— Tim & Julie Harris

📬 Thanks for reading Harris Real Estate Daily. Share this with a colleague who needs clarity about where the industry is headed.

Forwarded by a friend? Sign up with just one click here.

Share this issue.

GROW WITH EXP REALTY & LIBERTAS

You are implementing the systems and tracking the metrics (like we teach in the book, Harris Rules). Now you need a brokerage partner that is built for maximum leverage and accountability. Stop letting outdated brokers skim your commission and slow your growth.

If you’re ready to make 2026 your breakthrough year—by finally getting the systems, coaching, and commission split you deserve—it’s time to partner directly with Tim and Julie Harris at EXP Realty.

This is your mandatory next step to becoming a Millionaire Real Estate Agent.

👉 Go now to https://WhyLibertas.com/Harris or text Tim directly at 512-758-0206.