How to Work with a Seller Who Priced Their Home with AI

In this Member Training, Chris breaks down a situation agents are likely to run into more and more: a seller asking ChatGPT what their home is worth and then treating that answer like a real CMA. The session shows why AI can sound extremely confident while still working with incomplete, outdated, or misunderstood information. In one example, ChatGPT initially produced a polished valuation using public sources and “comps,” but further questioning exposed problems with the underlying data. Chris compares that with what an agent brings to the table through actual MLS data, carefully selected comps, property condition, local market knowledge, previewing inventory, and conversations with other agents.

More importantly, this training shows you how to handle the objection without arguing with the seller or telling them ChatGPT is wrong. Chris walks through a live example and a five-prompt sequence you can actually use with a seller to make AI question and improve its own analysis. The goal is to work alongside the seller, expose what the AI may be missing, introduce your real CMA data, and demonstrate why your experience still matters. You’ll also see another approach using the AI Studio Listing Coach, including language for validating the seller without agreeing, shifting the conversation toward actual buyer behavior, and getting back to a professional pricing strategy.

Helpful Timestamps

Times are approximate.

• ~3:00 – Why sellers are increasingly using AI to determine home values
• ~10:00 – Live example of ChatGPT attempting to price a home
• ~18:00 – Poking holes in ChatGPT’s original answer and checking the data behind the “comps”
• ~25:00 – ChatGPT vs. a real CMA: public internet data compared with MLS data, market knowledge, condition, and actual comp selection
• ~30:00 – The 4-step “ChatGPT Says” Objection Handler: Don’t dismiss it, ask what AI knows, show your analysis, and name the missing information
• ~35:00 – Live seller conversation and pricing objection case study
• ~43:00 – Asking ChatGPT, “How would you improve your analysis?”
• ~48:00 – Uploading a real CMA and letting ChatGPT reassess its recommendation
• ~53:00 – Adding property condition, remodeling, and information AI cannot know from public data
• ~58:00 – Using the AI Studio Listing Coach for another way to handle the objection
• ~67:00 onward – Member Q&A and additional implementation discussion

Put This Into Practice

1. Download the presentation slides and save the five-prompt sequence somewhere you can easily reach it before a listing appointment. The sequence starts with a simple valuation request, then checks comp dates, asks AI to critique itself, adds your CMA, and finally factors in the home’s actual condition.
2. Practice this before you need it. Pull up a property you know well and run the prompts yourself. The goal is to be comfortable enough that you could open your laptop or tablet during a listing appointment and walk through it with the seller.
3. Do not fight the seller’s AI. Work with it. Instead of saying, “ChatGPT is wrong,” let ChatGPT expose its own assumptions while you provide the information it was missing.
4. Bring your CMA. Your MLS data becomes the source of truth you can feed into the conversation rather than relying on whatever ChatGPT happens to find online.
5. Use the AI Listing Coach in AI Studio if you want help preparing for this objection, pricing conversations, seller questions, or other buyer and seller situations.

Recommended: Watch the full training rather than just grabbing the prompts. The real value is seeing how Chris introduces the questions without putting the seller on the defensive and turns an AI-generated objection into an opportunity to demonstrate expertise and build trust.