Check out our new The REAL AI Podcast for this week – Episode 52 11/9/26 – AI-generated with hosts Miles and Grace
REAL AI: The ChatGPT hack nobody teaches, Thanks NAR NXT, AI stats and facts, headlines, and AI quote of the week
By Kevin Hawkins with Korey Hawkins | Nov. 9, 2026 – Vol. 4 Issue 45
REAL AI is the industry’s first and most popular weekly newsletter (and podcast) helping real estate professionals stay on top of the most important AI developments to save time and stay safe. Now bigger and better than ever, the best-selling real estate book The REAL AI Guide for Real Estate Agents is available today on Amazon!
The ChatGPT hack nobody teaches
Have you ever been working with ChatGPT Business or Pro in a long chat or thread, and suddenly, it seems to get stupid? It forgets something you already told it, repeats a mistake you already corrected or starts giving you answers that make you wonder if it’s even paying attention.
You’re not imagining things. And here’s the surprise: The longer you work in the same ChatGPT conversation, the more likely you are to run into problems like these.
Most of us assume the opposite. We figure the longer we work with ChatGPT, the better it gets to know us and the more useful it becomes. That makes sense, right? Unfortunately, that’s not always how AI works.
The good news? There’s a remarkably simple ChatGPT hack that can rescue your work without having to start from scratch.
What to do when ChatGPT starts getting stupid
ChatGPT has something called a context window. Think of it as the amount of information ChatGPT can actively keep track of while answering your questions. It’s enormous when compared with what we had just a couple of years ago, but it’s not unlimited.
When conversations become extremely long, and especially when you’ve made multiple corrections and refinements, earlier information may get summarized, compressed or pushed outside that active window. That’s when ChatGPT may start forgetting things.
Even if you already covered it within a massive thread, ChatGPT can’t always consider everything you’ve covered. So when you notice the quality of responses slipping, especially after working in the same thread for a long time, you need to consider a fresh start.
This is when you ask ChatGPT if it’s time to start a new conversation. Ask: “Is it time to start a new thread, or should we keep going?” If ChatGPT recommends starting fresh, ask it to prepare everything you need to move to a new conversation without losing the important details of your work.
Don’t start over. Start fresh.
Here’s what to do next. Instead of abandoning your conversation, ask ChatGPT to create a complete handoff document. Think of it as creating a full brief for a colleague taking over when you take a vacation.
It can help capture everything important you’ve accomplished. The document it prepares allows you take that information into a brand-new conversation.
The ChatGPT conversation rescue prompt
Here’s a prompt you can copy and paste directly into your existing ChatGPT thread:
“I want to continue this project in a new ChatGPT conversation without losing important context.
Create a comprehensive handoff document that includes:
- The project’s purpose, goals and current status.
- All important decisions, facts, figures and conclusions we’ve established.
- My instructions, preferences, writing style and corrections that must be preserved.
- Any essential documents, sources, links or files, clearly identifying which need to be uploaded again.
- What we’ve completed, what remains unfinished and the next recommended step.
- Any unresolved questions, conflicting information or details you’re uncertain about.
Separate confirmed facts from assumptions. Do not invent missing information. Format everything as a self-contained handoff I can paste into a new conversation. End with a suggested first prompt for that new conversation.”
Next, practice HER (Human Evaluate and Review). Once ChatGPT creates your handoff, review it to make sure it hasn’t missed anything important. Then start a new chat, paste in the handoff and any other documents ChatGPT recommended you upload.
You’re back in business with a fresh conversation that can deliver more accurate answers without starting over.
But what about ChatGPT’s memory?
Memory can help ChatGPT remember useful information about you across conversations and projects, but memory is not a perfect transcript of everything you’ve ever told ChatGPT.
That’s why this hack is so valuable. You’re not depending on ChatGPT to magically remember every detail. You’re giving it a written set of instructions to pick up where you left off. The goal is to improve the accuracy of its answers.
Why real estate agents should care
You are in the trust business. The last thing you need is for ChatGPT to forget key details when it’s assisting you in preparing something for a buyer or seller. Or worse, give you a lousy answer.
It’s also why the HIM (Human in the Middle) and HER (Human Evaluate and Review) principles still apply. Check your handoff for accuracy and don’t wait until ChatGPT starts acting stupid to create a handoff.
The bottom line? When ChatGPT starts forgetting, don’t start over. Start fresh. (-Kevin)
Need a standing-room only AI class for your agents or staff? Or a veteran moderator for your AI panel? How about an AI keynote on “The Good, The Bad and the Ugly of AI?” Hire WAV Group’s Kevin Hawkins, The REAL AI Guy and Amazon best-selling author of “The REAL AI Guide for Real Estate Agents” – the 3rd edition just released! How to hire Kevin here.
AI Facts + Stats
1. 70% of real estate agencies surveyed now use AI for personalized email marketing.
2. 60% of real estate developers surveyed use AI to analyze macroeconomic indicators (e.g., GDP, interest rates) to inform project decisions.
3. 50% of top real estate agents surveyed use AI-powered video tours.
4. 73% of real estate buyers surveyed use AI chatbots to answer property-related questions.
5. AI-powered translation tools (e.g., Google Translate integration) are used by 60% of real estate firms targeting international buyers.
Source: WorldMetrics (-Korey)
AI Headlines
Unlicensed AI bots are closing real estate deals, industry leader warns | 11/3/26 Inman
Controlling rouge AI in real estate starts with licensing and human supervision.
AI is changing home listings. Can you trust what you see? | 11/4/26 Clickorlando
Buyers and regulators are growing more concerned with AI’s use in listing images.
Voters Rejected A.I. in the Midterms. Now Comes an Era of Scrutiny. | 11/4/26 The New York Times
Data center siting, energy use and AI regulation are becoming major political issues after years of limited oversight.
Five remarkable trends from NAR’s 2026 Profile of Home Buyers and Sellers | 11/5/26 WAV Group
The internet is still king, but AI search is coming.
Wanted: AI Babysitters | 11/4/26 Korn Ferry
Banks are hiring staff to review AI output, manage compliance and oversee human-AI interactions.
How AI Is Compressing Real Estate Decision Timelines | 11/4/26 Propmodo
Faster underwriting and portfolio analysis give firms an edge when speed and certainty matter.
Flexmls Accelerates Real Estate AI With the Launch of Scout | 11/4/26 RISMedia
Users can query authorized MLS data, build CMAs and check market statistics in natural language.
Economists Question How Long AI Can Fuel Industrial Real Estate Demand | 11/3/26 GlobeSt
Investors face growing questions over whether AI spending can generate enough returns to sustain infrastructure demand.
CrowdStrike Looks To An AI Kill Switch To Beat Rogue Agents | 11/3/26 Forbes
The new control can stop unauthorized agents in real time, but it cannot undo actions already taken. (-Korey)
AI Quote of the Week – Matt Consalvo, Former CEO of ARMLS, as told to Real Estate News
Want a signed new edition of our Amazon best-selling The REAL AI Guide for Real Estate Agents? Here’s the secret link! Story ideas? Email korey@wavgroup.com.