Clayton Smith • YouTube • October 2025
Overview
This is the full recording of Clayton Smith's keynote at Mogul Mastermind on September 24, 2025. The talk was built for real estate investors and entrepreneurs who are curious about AI but haven't yet committed to a consistent practice. Clayton's core argument is straightforward: AI is not magic, not a job-killer, and not something to be mystified. It is a pattern-predicting tool that multiplies your time and your creative output -- if you know how to direct it.
The presentation moves through what AI actually is, how prompting works, a tour of the major platforms, and a range of practical use cases grounded in Clayton's own production and consulting business. The session closes with a live audience demo: a ChatGPT voice-prompted real estate ad built in real time from crowd-sourced property details.
The Q&A section is as valuable as the presentation itself. It covers privacy, platform biases, AI for contract review, condo document analysis, automated voice agents, deal underwriting, and the longer-horizon question of what happens to repetitive work as the technology matures.
A free prompting guide and resource list are available at claytonsmith.ca. Clayton also introduced his Virtual Mastermind custom GPT -- a Napoleon Hill-inspired advisory council distilled into a shareable tool -- and offered it directly to the audience via QR code.
Why This Matters
Most AI content falls into two failure modes: either it hypes AI as a civilization-ending force requiring urgent adaptation, or it stays so surface-level that the audience leaves with nothing actionable. This talk threads that needle. It is grounded in real workflows, delivered by someone who bills clients using AI daily, and validated in real time with a live demo. That credibility matters.
For real estate investors specifically, the concrete use cases presented here -- contract review, condo document analysis, deal underwriting support, content generation, and property listing copy -- represent meaningful time savings at low cost. The barriers are lower than most people assume. You do not need technical skills. You need a clear question and the habit of asking it.
The broader framing -- from doers to directors -- is the most durable idea in the session. As AI absorbs repetitive, template-driven tasks, the human role shifts toward judgment, strategy, and creative direction. Entrepreneurs who build that muscle early will have a structural advantage. Those who wait will find the gap harder to close.
This page captures the talk for future reference and as a training and onboarding asset. The prompting guidance, tool landscape, and use case inventory will remain relevant well past the event date.
Metadata
Key Points
Quotable
AI-generated from source material. Verify important details against the original source.
Clayton Smith
"It is a multiplier of your time and for your creativity. You are still going to be important in this process."
The clearest single-sentence reframe of what AI actually is. It sidesteps both the hype and the fear, and puts the human back at the center of the equation. Useful for any client or audience new to the technology.
Clayton Smith
"Think of it as kind of an intern. You have to be very specific with what you're giving -- with what you're asking of AI -- and you have to be specific with what you're looking for."
The intern analogy is the most durable mental model for prompting. It immediately sets accurate expectations and gives new users a concrete framework for structuring requests.
Clayton Smith
"It is going to allow us to go from doers to directors."
Compact, quotable, and strategically important. This line reframes AI adoption not as a threat but as a role upgrade for entrepreneurs willing to develop new skills.
Clayton Smith
"AI is just a tool. It's actually a tool that can build a lot of other tools. It's not something to be mystified. It is something to get familiar with."
Grounds the conversation in practical utility rather than speculation. The second sentence -- that AI can build tools -- is a subtle but important point that separates experienced users from beginners.
Clayton Smith
"I got Chat GPT to teach me step by step how to do all of that stuff. And you can all do that, too."
A direct, credible proof point. Clayton learned animated map production in six months through AI instruction alone. It demolishes the "I'm not technical enough" objection for anyone in the room.
Concepts and Ideas
Core Frameworks
AI as Pattern Predictor, Not Oracle
AI generates outputs by predicting the most statistically likely continuation of a pattern based on its training data. It does not reason, verify, or understand. This means it can be confidently wrong. The practical implication is that AI works best as a drafting and brainstorming engine -- not as a fact source. Anything it produces that matters should be verified against primary sources before acting on it or sharing it.
The Intern Mental Model
Thinking of AI as a smart intern with broad general knowledge sets the right expectations. A new intern does not know your business, your preferences, or your standards. You have to give them context, clear instructions, and specific outputs. The better your briefing, the better the result. Vague inputs produce vague outputs. This model also removes the intimidation factor -- most people already know how to delegate to a capable but unfamiliar assistant.
From Doers to Directors
As AI absorbs repetitive and template-dependent tasks, the human role shifts toward creative direction, judgment, and strategy. For entrepreneurs, this is an expansion of leverage -- a single person can now direct the equivalent of multiple virtual workers across writing, analysis, research, and coding tasks. The transition requires building comfort with delegation and output review rather than hands-on execution.
Prompting Principles
Specificity as the Core Skill
The quality of an AI output is almost entirely determined by the quality of the input. Generic prompts produce generic outputs. Adding context, role framing, output format, and explicit constraints produces outputs that are useful on the first pass. Clayton's rule: treat every prompt the way you would brief a capable contractor who has never worked with you before.
The Three-Question Technique
Adding "ask me three questions, one at a time, to refine your answer" to a prompt transforms a static request into a short interview. The AI draws out context that the user may not have thought to provide, then produces a more accurate and tailored output. The "one at a time" constraint is critical -- without it, AI will front-load a cascade of questions that is harder to process and respond to.
Prompt Refinement via AI
You can ask AI to generate a better prompt from rough inputs. Dump your context and objectives in plain language, ask it to produce a well-structured prompt, and then use that prompt in your next session. This is especially useful when you know what you want but struggle to articulate it precisely. AI's strength at pattern recognition extends to recognizing prompt structures that produce good results.
Real Estate Applications
Contract Review as a Flagging Tool
Feeding a contract into ChatGPT does not replace legal advice, but it surfaces clauses you might miss in a quick read. Clayton cited a real example where AI flagged a buyer-inserted clause that shifted legal fee payment to the seller -- a detail he would otherwise have brushed past. The recommended use: ask AI to summarize key terms, flag unusual clauses, and identify who is responsible for specific obligations. Use the output to ask better questions of your lawyer, not to replace the lawyer.
Condo Document Analysis
Condo documents -- disclosure packages, bylaws, meeting minutes -- are dense and easy to skim past. AI services that analyze these documents (many of which use AI internally) can perform a meaningful preliminary pass to flag rental restrictions, short-term rental prohibitions, unresolved maintenance issues, and signals of upcoming special assessments in the minutes. This does not eliminate professional review, but it front-loads the risk identification before money changes hands.
Content Generation for Listings and Marketing
The live demo illustrated how quickly AI can produce a full property listing draft from a handful of facts and a tone direction. The output handles structure, vocabulary, and flow -- tasks that slow most people down. The human role is to provide the inputs, direct the tone, review for accuracy, and refine from there. AI handles the donkey work; the operator handles quality control and brand voice.
Platform Landscape
Platform Selection and Bias Awareness
ChatGPT, Grok, Gemini, Copilot, and Claude are all trained on different data sets with different weighting and priorities. This produces different outputs, blind spots, and biases. Some advanced users cross-check outputs between platforms -- feeding output from one into another for a different angle. Grok has an advantage in real-time information via X integration. Gemini integrates with Google Workspace. ChatGPT remains the most general-purpose and widely used starting point.
Custom GPTs as Reusable Instruction Packages
A custom GPT is a version of ChatGPT pre-loaded with your own system instructions, context, and behavioral rules. It behaves consistently across sessions without re-prompting. The practical value: package specialized knowledge, a persona, or a repeatable workflow into a shareable tool. Clayton's Virtual Mastermind GPT -- modeled on Napoleon Hill's advisory council concept from Think and Grow Rich -- was built this way and shared directly with the Mogul Mastermind audience.
Observations and Warnings
The Red Line: Fabricated Realism
AI image and video generation tools will invent content that does not exist when pushed past the edges of source material. Clayton's example: an AI video tool, asked to animate a still real estate photo with simulated drone movement, fabricated a non-existent subdivision when panning beyond the original image boundary. For any application where accuracy and truthfulness are required -- real estate marketing, legal documents, investment analysis -- this is a hard limit. Fabricated realism is not a quirk; it is a liability.
Privacy Defaults Are Not Protective
AI platforms store conversations. ChatGPT previously surfaced user conversations in public search results. The default settings on most platforms allow conversation data to be used for model training. Users on ChatGPT Plus can disable this, but it requires a deliberate settings change. The practical rule: treat AI sessions like a public forum. Anonymize names, generalize sensitive details, and never paste raw client data, contracts, or identifying information without thinking about where it might end up.
Implementation
AI-generated from source material. Verify important details against the original source.
Start a Daily AI Habit Today
Replace at least one Google search per day with a ChatGPT query. The goal is not efficiency -- it is familiarity. Fluency with AI comes from repetition. Ask it things you already know the answer to. Compare its output to your expectations. Build the instinct for when and how to use it before you depend on it for anything that matters.
Learn the Three-Question Prompt Pattern
Add "ask me three questions, one at a time, to refine your answer" to any request where you want a tailored output. Use this for ad copy, business strategy, deal analysis framing, or content ideas. Practice it until it becomes automatic. This one habit will improve your output quality faster than any other single technique.
Run Your Next Contract or Offer Through ChatGPT
Take a purchase agreement, lease, or offer you are reviewing and paste it into ChatGPT. Ask it to: summarize the key obligations of each party, flag any unusual or potentially unfavorable clauses, and identify who is responsible for legal fees and other costs. Use the output to prepare sharper questions for your lawyer -- not to replace the review. Anonymize names before pasting.
Generate Property Listing Copy with a Structured Prompt
Provide ChatGPT with property facts (size, rooms, features, location, nearby amenities), then add tone direction (luxurious, family-friendly, investor-focused) and ideal buyer profile. Ask it to draft ad copy, then refine from there. Use the voice feature on the mobile app if you are walking a property -- dictate the details in real time and have a draft ready before you leave.
Use AI as Your Skills Instructor
Identify one technical skill you have been putting off learning -- video editing, GIS mapping, Python scripting, spreadsheet automation, anything. Ask ChatGPT to teach you step by step. Start with "explain what I need to know to get started" and build from there. Clayton learned animated map production this way in six months with no prior experience. Pick your skill and start the first session this week.
Build or Access a Custom GPT for Repetitive Workflows
If you have a task you do repeatedly -- drafting follow-up emails, analyzing deal spreadsheets, writing social content in a consistent voice -- build a custom GPT with your preferred instructions pre-loaded. Scan the QR code from the presentation to access Clayton's Virtual Mastermind GPT as a starting example. Study how it is structured, then build one for your own use case.
Lock In a Privacy Protocol
Go into ChatGPT Plus settings and disable conversation data from being used for model training. Establish a standing rule: never paste raw client names, contract details, or identifying information. Use placeholders ("the buyer," "the property at address X," "my client in the tech sector"). This takes five minutes to set up and prevents a class of problems that is embarrassing and potentially expensive.
Explore Otter.ai for Meeting Capture and Content Mining
Use Otter.ai to transcribe your next meeting, call, or presentation. Feed the transcript into ChatGPT and ask it to extract key insights, action items, follow-up questions, and ideas for content. This converts conversations that would otherwise be lost into structured intelligence and reusable material. Start with a single recorded meeting this week.
Download the Free Prompting Guide
Clayton's prompting guide and resource list are available at claytonsmith.ca via the email opt-in. Download it, read through the prompt frameworks, and pick two or three to test immediately. The guide covers prompt structures, platform recommendations, and techniques for getting consistent outputs across different use cases.
Tools & Resources
Some links on this page may be affiliate or referral links. If you sign up or purchase through them, there may be a mutual benefit -- at no additional cost to you.
Mentioned Resources
| Resource | Description |
|---|---|
| ChatGPT | Clayton's primary AI platform. Used for scripting, brainstorming, coaching, coding, and general-purpose tasks. ChatGPT Plus includes a training data opt-out toggle in settings. |
| Grok | Clayton's secondary platform. Tied to X (Twitter) for real-time information. Used as a fallback when ChatGPT is slow and for cross-checking outputs. Free tier available. |
| Gemini | Google's AI platform. Comes with Android devices and integrates with Google Workspace (Sheets, Docs, Gmail). Audience members noted meaningful limitations and errors in its current state. |
| Microsoft Copilot | Microsoft's AI assistant, widely understood to be built on ChatGPT's underlying model. Integrated into Microsoft 365 products. |
| Otter.ai | AI transcription tool Clayton uses to capture meetings and presentations. Transcripts are fed into ChatGPT for insight extraction and content development. Paid subscription. |
| ElevenLabs | AI voice and audio platform. Clayton uses it for voiceover production in real estate videos. Also capable of voice cloning and music generation. Paid subscription. |
| Sora | OpenAI's video and image generation tool, available to ChatGPT users. Used to generate AI art and short video clips. Note: will fabricate content beyond source image boundaries. |
| Ballpark Housing | Locally developed (Edmonton) AI-powered property analysis tool built by Akeem Brown. Designed to speed up underwriting for property developers. |
| claytonsmith.ca | Clayton's website. Offers a free download of his AI prompting guide and resource list via email opt-in. Also covers his media, marketing, and AI consulting services. |
Suggested Resources
| Resource | Description |
|---|---|
| Claude (Anthropic) | Mentioned by audience members as a strong coding and document analysis platform. Particularly noted for handling long documents and nuanced instruction-following. Complementary to ChatGPT rather than a replacement. |
| Perplexity AI | A research-focused AI platform that cites sources inline. Useful for verifying AI outputs and for tasks where traceability of information matters -- contract research, market data, regulation lookup. |
| Think and Grow Rich -- Napoleon Hill | Referenced directly by Clayton as the inspiration for his Virtual Mastermind GPT. The concept of an advisory council of historical mentors maps naturally to custom GPT persona design. |
| ChatGPT Custom GPTs | OpenAI's official documentation for building and sharing custom GPTs. Starting point for packaging your own instructions, knowledge, and workflows into a reusable tool. |
| RECA -- Condo Document Guide (Alberta) | The Real Estate Council of Alberta's guidance on what to look for in condo disclosure documents. Use alongside AI document review to ensure you are asking the right questions. |
Source Material
[00:01]
Now ladies and gentlemen, it's time to move beyond the buzzwords and hypes surrounding artificial intelligence and start using it as a real tool for growth. Our speaker is a fellow entrepreneur and the perfect guide for this journey. He's an expert at cutting through the complexity and showing other business owners how to get tangible practical results from AI. So tonight, he's going to show us exactly how to leverage these tools for efficient contact creation, smarter deal analysis, and automating the repetitive tasks that eat up our valuable time. Get ready for a session packed with actionable strategies that you can implement tomorrow. So please put your hands together for our keynote speaker and welcome Clayton Smith. [Applause] All right. Thank you very much, Tracy. Good evening, everybody. It takes a lot of work to put on a presentation like this. I'm not talking about my presentation, but just the whole event here. And it is great to have a forum where we can all come together and talk about ideas, talk about making money, talk about bettering our lives. So, I wanted to give a hand to Devin, Tracy, and the Mogul team here tonight. So please join me in giving them all a hand.
[01:30]
And also, of course, a hand for the people online there. So we see you. Give them a hand too. All right. So, I wanted to talk to you folks here tonight about AI. Who here is dabbling with AI? You're playing with AI or you've got a ChatGPT account? How many here have ChatGPT right now? Okay, good. Not everybody. Who does not? Okay. So a couple of you. How many of you are paying for ChatGPT right now? Okay, almost the same. All right, so what I was going to cover here tonight is just a little bit of what AI is and what AI is not. Try to demystify things a little bit for people that are just wanting to get into it. They're wanting to find out how to use AI, how to make their lives a little more efficient, and how to maybe get through some deals quicker, make a little bit of money with it. There's a lot of ways that I use AI today and I'll get into some of that. We'll get into some prompting techniques because prompting is kind of how you get the juice out of AI. So we'll go through a little bit of that and then I've got a few things you can download here. I'll tell you about those towards the end of the presentation.
[02:30]
So, a little bit about what AI is and what it isn't. It is basically a pattern predicting tool. It's a multiplier of your time and for your creativity. So you are still going to be important in this process here. You're hearing a lot of talk these days about AI just taking everybody's jobs over. It's going to put everybody on the unemployment line. I don't know what's going to happen, but right now it's a way to kind of leverage yourself and your own capabilities. That's probably the biggest thing I would say you can think of in terms of AI. Some things to be aware of: AI isn't always accurate. So for those of you that have played with ChatGPT, have you ever tried to give it some math problems? How does it do? Yeah. So you've really got to be careful with some of the information that comes out of it. You're still going to need to verify a lot of things that come out of AI. So you want to make sure that you're not just copying and pasting and definitely not taking this information to the bank without a little bit of verification.
[03:30]
All right. So just a few examples of how I personally am using AI today. I use AI a lot in my own production business. So a lot of what I do is I produce a lot of videos for real estate projects and I use AI for the full basically for the full production process. So I use it to basically start writing a script. I use it to catalog the videos that I have to put together the script with. I use it to generate voiceovers for the videos. So I can show you the platform I use for that. I use it for coaching, for teaching. So for teaching special techniques on how to do things. There are these animated maps that I create as part of my process. I didn't know how to do that about six months ago. I got ChatGPT to teach me step by step how to do all of that stuff. And you can all do that too. So if there's anything you want to know, ChatGPT can be your instructor to learn whatever technique it is that you want. Oh, a couple of other things there -- brainstorming, business strategy. I've got a tool that might help you do that later. So we'll get into that. Coding websites. I've built a website, built a few websites. I built a few online tools. I built tools that I actually use in my business to make my life a little easier.
[05:01]
So there are a few platforms that are out there. These are some of the most popular AI platforms that are available right now. ChatGPT -- most people have heard of it. How many people here use Grok? One person. Grok is actually my backup. I don't have the paid subscription for it, but when ChatGPT sometimes gets a little bit backed up, that's when I switch over to Grok to carry on with whatever I want to do. Gemini is a popular one here. Gemini comes with -- if you've got an Android -- Gemini comes with every Android device now. So you've already got it on board there. Especially if you've got a Gmail account, you've got Gemini. Copilot from Microsoft. I've heard somebody say, and I think this is true, that it's basically Microsoft's white label for ChatGPT. I actually don't like Copilot for some reason. I don't know why. It just -- the first time I used it about a year ago -- it just didn't give me what I was looking for. It was very basic and just wasn't impressive. So I haven't used it much.
[07:00]
GPT stands for Generative Pre-Trained Transformer. There you go. I went through this so many times this weekend when I was rehearsing this. All right, we'll get into prompting a little bit here. Prompting -- if you can think of AI as an assistant with basic knowledge, you have to train it and it doesn't read your mind. So think of it as kind of an intern. You have to be very specific with what you're asking of AI and you have to be specific with what you're looking for. One of the things I like to do is I like to have the AI interview me. So one of the things I'll add to my request -- I'll say, "Ask me three questions, one at a time." That's important. Otherwise it'll just avalanche you with a bunch of questions. But ask me three questions one at a time to refine your answer. And usually it does that. Usually it'll just ask you one question, it'll go on to the next, and it'll go on to the next. AI can also refine your prompt. You can actually ask AI to give me a good prompt. You can basically throw a bunch of information at it, say "I want a good prompt from this," and then it'll give that back to you.
[10:00]
Some tools that I use. Otter.ai. This is one that I use a lot. I use this for transcribing meetings. So if you have an online meeting, you have an online call, or if you have a -- like I've got a microphone here -- I'm recording myself as I'm doing the presentation. I can take the transcription from this from Otter.ai and then I can use that later on by feeding that whole transcript into ChatGPT and then extracting some insights and some ideas out of that for posts. ElevenLabs. That is the platform that I use for producing voiceovers. What is neat with this -- you can produce music with this. You can even duplicate your own voice with this platform. So if you ever thought of creating an audio book and you just thought, "Oh my god, I can't sit for 14 hours and just read this book out" -- well, you just put the whole thing into ElevenLabs and it'll just read it out for you. There's actually an audio book that I have that I bought recently and the author said that he actually made the whole thing with an AI voiceover. You could tell because there were little things in the tone -- a few little items where something sounded like a question but it wasn't said as a question so it doesn't quite get the tone right. And there was another part where another speaker started to adopt an Australian accent. So that was a little odd.
[12:00]
Sora -- that is, if you're a ChatGPT user, you might be familiar with Sora. Sora is a video and image generating platform and you can use that to create some very impressive art, lifelike images, and also short video clips. I've done that a few times and it's pretty neat what you can do when you start getting into this. Some advanced resources especially for real estate investors. Custom GPTs. What a custom GPT is -- it's basically a customized AI. It's you could either look at it as a reusable prompt, but it is basically your own GPT that is programmed with your own set of instructions. You can tweak it, you can refine it, and you can keep on using it and it'll keep on delivering fairly consistent outputs. Ballpark Housing -- ballparkhousing.com. It is being developed locally here. Akeem Brown actually is developing that site. It is powered by AI and it uses AI to really speed up the analysis process for property developers. Builder Chat -- there are some guys that I was introduced to in the last few weeks and they're building a service that helps builders to basically retain leads and increase conversions. They can install an assistant on a website that lives on that website 24/7. It is not a chatbot. It is an AI-powered interactive tool that can have an ongoing interactive conversation with a prospect and then it'll be able to help the sales team with delivering that prospect and hopefully turn them over into a sale.
[14:00]
How many here have heard of automated voice agents? A few people. Not too many though. Have you tried any of them? They've come a long way just in six months. So what an automated voice agent is -- it's basically a live AI-powered agent that you can speak with. You can have a phone conversation with. The voice is totally AI generated. The responses are powered by AI. It can be difficult to tell if you're talking to AI or if you're actually talking to a human being. Where you get a hint that you're actually talking to an AI -- sometimes there are pauses that are just a little longer than what would be natural. Other than that, the voice is pretty clear.
[15:01]
What AI is going to allow us to do -- it is going to change how we do everything. Five years from now, we're not going to be doing a lot of the things we do now. When it comes to any kind of a repetitive task, any job that relies on templates, a lot of that is going to get automated. When it comes to things like generating content, generating ads, things like that, we're going to be using AI for that a lot. And we're just going to be able to send a task to an AI assistant of some sort. What we're going to be doing as creative people, as salespeople, as entrepreneurs, we're going to be basically directing virtual employees. So it allows us to go from doers to directors. So that's going to free up a lot of time for some people. For some people, it's going to free up more time than they like. Because some people are going to have to learn how to do things a little bit differently. Their repetitive jobs are going to probably go by the wayside. If you think about it -- think about things like receptionists in some ways. We were just talking about virtual voice agents, right? The technology has come so far in the last few months to make it almost indistinguishable. You think of where that's going to be in about another year and a half, two years, right?
[16:30]
In the big picture, AI is just really just a tool. It's actually a tool that can build a lot of other tools. So it's not something to be mystified. It is something to get familiar with. And what I would suggest is to start getting into a daily habit of using AI every day. And that means -- how many people are still doing Google searches? How about instead of using Google, use an AI platform. Whether it's ChatGPT, whether it's Grok, use that instead. Just to start getting used to it because it is going to replace Google. Google is pretty well going by the wayside as it is anyway. AI is able to come up with different answers and in some ways better answers. I rarely use Google anymore.
[18:00]
Audience member: One of the things I did want to add as to why it is better to go to something like ChatGPT over Google search is that when you search something, keep in mind that the top stuff that comes up has been paid for. So there's a huge bias. Whereas if you go to ChatGPT at this point there is nothing that's paid to have that thing pop up. So you're going to get a completely unbiased recommendation as opposed to a biased one on Google's.
[19:00]
Clayton: Right. As far as I know, there's no paid way to elevate searches on AI platforms like through ChatGPT. Now it depends -- I just saw here in the last few weeks that it's interesting how ChatGPT has been trained. Apparently it's using a lot of Reddit and Wikipedia. So we do want to be careful with that. That's why it's important to verify the information you're getting out of these. So if you want help creating some kind of report or a proposal or anything like that, AI is good right now for doing a lot of that heavy lifting -- actually just grinding out the paragraphs. But you've got to take the time, go through it, and review the information that's in there to make sure that if you're passing this information on to somebody else, this information is good. So you still have to verify it at this point. Now give it a few years and the technology will evolve so much so fast that a lot of that is probably going to get taken care of -- you're going to be able to trust the technology. Right now you're just not quite able to trust that technology yet. Not enough to sign off on it.
[20:02]
All right. So that kind of concludes the formal part of the presentation here. There is a QR code up there that takes you to my website. There is a document that you can download there that has a prompting guide and then there's a list of resources that you can refer to there. The other thing that I wanted to mention -- if you want help with things like creating custom GPTs, implementing AI into your business, I can help you with that too on top of some of the media services that I offer. But you can go to my website there. It's got my links and you can connect with me that way on social. Other than that, we'll take a few questions. I'll take a couple of questions and then I wanted to do a demo on using ChatGPT here.
[21:30 -- Q&A]
Question (privacy): How do you deal with the privacy aspect of it in terms of putting information in or if it recreates your voice? How do you protect anybody else from that? Clayton: On the privacy aspect -- one of the things you've got to keep in mind is that whenever you're dealing with any kind of a digital platform like this, that information is getting stored somewhere and at some point somebody could very well get access to it. Actually there was found out that a bunch of conversations were put on Google and available for searching -- and people were using ChatGPT like a psychiatrist and they were putting some very personal bits of information in there. People are putting contracts in there and things like that. So you want to make sure that when you're using ChatGPT, you want to kind of anonymize things a certain way. Instead of saying "I have this issue," maybe you might want to say "a partner or a client has this issue" and try to avoid using personal names. Just assume that this information could well get out there. So you want to make sure you're treating especially client information with a lot of sensitivity because you don't want a situation where your client's information gets out onto the internet -- that could be a bit of an embarrassing problem for you and expensive. With ChatGPT Plus there is a toggle that you can switch off so that your personal information isn't used to train the models. So there is a little bit of a privacy benefit to that. But it's not by default. You have to go in and actually flick that switch over.
[25:00 -- Q&A continued]
Question (AI courses): I have a friend who sees a lot of courses about learning AI in the next 30 days or 60 days or whatever. How do you tell the good tools from the bad tools? Clayton: I think a lot of those -- I've taken a couple of them -- I think they all have some good information and a lot of it is available out there for free anyway. And ChatGPT can teach you a lot of this itself. Of course it all comes down to the implementation. It still comes down to you. You've got to hire the right resources, do it. You still got to put all that stuff into action. So the course -- just like how many people here have taken a real estate course -- is taking a course just enough to make money in real estate? No. So it's the same kind of idea. There's some good information in there. There's some good techniques. A lot of times, especially with the lower level courses, they give you just enough to upsell you on to the next level, right? From the $27 course to the three-day boot camp to the $10,000 coaching program.
[35:03 -- Q&A continued]
Question (platforms): You mentioned ChatGPT a lot and we all know about it. But you also mentioned about other ones for video and arts and everything. Is it important for us to know each and everyone for specificity or does ChatGPT cover it all? Clayton: ChatGPT is kind of my Swiss army knife. I can get by without the other platforms. The other platforms can be good for some different things. I've heard that some people like using Grok because it's tied in with X and so there's a lot of real-time information there. Some people have said that Claude is another AI platform that's out there. A lot of people like using that for coding, for creating, for developing applications. Funny enough, a friend of mine and I are actually developing an automatic trading system. We spent about three hours on a call and then my friend took the ball and ran with it. He went and used ChatGPT to come up with the code -- and he's not a programmer. He put together a rough working program. So it was actually doing a few things but it would just stop. He hit a wall. I said "try Claude -- I know that people have said they like using Claude for coding." So he did that and he said that actually was worse. So I thought, okay, try Grok. He texted me back about three hours later and said "I think we actually have something that works." He's got it in testing right now doing some paper trading -- tying TradingView to Interactive Brokers and executing automatic trades. Pretty neat stuff. Different platforms will do different things. ChatGPT is a really good general-use AI platform. Like most other things, people are going to like the first thing they start with. So if you start off with Grok or Perplexity or any of the other platforms out there, if that's the one you started with and you just get used to how it works, you'll probably just stick with it.
[37:30 -- Q&A continued]
Question (platform biases): Have you noticed particular biases of certain platforms? When you ask the same question across different platforms, there are some huge biases. Clayton: I've heard people say that. I haven't looked out for it myself so I can't really say it, but I've heard plenty of people say that -- yeah, there are different biases out there. Especially with Grok -- there was a famous incident where it put out some pretty interesting information when it came to a certain former German chancellor. But that apparently got fixed pretty quick. They're all trained on different sets of information. So they all have their own little tweaks and their own sets of information they're trained with. I have heard people say that one thing they like to do is they take the output from one platform and then they feed it into another and have it reanalyze it. Because they'll get a slightly different viewpoint on each one of them and a different angle, and it'll a lot of times come up with ideas you just didn't think of going in.
[41:00 -- Q&A continued]
Question (memberships): Do you have memberships with several of the platforms you've mentioned? And how different are they? For those in the room thinking about getting a membership -- curious on the significant difference. Clayton: I pay for ChatGPT. I pay for Otter. I pay for ElevenLabs because I'm using those in production work. The rest of the platforms I've got free accounts with them. One thing I will mention with ChatGPT -- with ChatGPT Plus there is a toggle that you can switch off so that your personal information isn't used to train the models. So there is a little bit of a privacy benefit to that. But you have to go in and actually say you want to keep your stuff private -- it's not the default.
[42:00 -- Q&A continued]
Audience member (contract and email review): I plugged in one of my contracts to analyze. I never even thought to remove personal information. But it was definitely super helpful because it was like -- what are the requirements in this contract? The other thing too is action items that we might have missed in conversations. We had an email thread going back and forth and then using that to say "what were the talking points in this email?" And I found one time that's where Gemini was useful -- it was able to pick up the things the other person was asking. I was like "oh, I missed that." When I read through the email I missed that thing, but then Gemini was able to go through it and pick up on the asks and the commitments made. Clayton: And that's a good one. Now I'll say that's okay for a quick review of something. You again do not want to trust the recommendations that come out of it just without question. It might be able to help you get some insights out of the contract that might help you to ask more informed questions of your lawyer. There might be little things in there, little clauses that you may not have just brushed over them before. I've run a few contracts through ChatGPT and it actually flagged a few things for further consideration where I just thought -- oh, okay, I didn't actually think of it that way. So one was a lease agreement and one was a recent offer to purchase that I received on a condo. I missed the clause where the buyer had put in that the seller pays the legal fees. Little things like that.
[45:03 -- Q&A continued]
Audience member (condo documents): I know that when purchasing a condo, you've got to have a condo doc review. Some people do it themselves. Some people hire a company and some of those companies are starting to use AI to review the docs. Clayton: Yeah. You can take those condo documents and there are services that will analyze these for you -- they're probably going to be using AI anyways. But this is a way to do a good preliminary pass of those documents. At a high level, flag some items that might be in those condo documents. There might be restrictions in there -- you might not be able to rent it, you might not be able to rent it short-term. Things like that. There might be some items in the minutes that might flag you to the thought that there may be a special assessment coming up because there is a major maintenance issue that somebody was aware of and it might be an issue when the next report comes out.
[47:00 -- Live Demo]
Clayton runs a live demo using the ChatGPT mobile app. He prompts it by voice: "We need you to help us write a real estate ad. We'll give you some facts. Then I want you to ask me three more questions before proceeding with the draft. Do you understand?" ChatGPT responds: "Absolutely. I understand perfectly. Go ahead and give me the facts you have. And once I have those, I'll ask you a few clarifying questions before we jump into drafting the ad." Clayton (with audience help) provides: 2500 sq ft, Park View neighborhood, 4 bedrooms, triple car garage, golf simulator in the basement, across the street from a school, two blocks from a strip mall, three blocks from LRT, master suite with a large jacuzzi tub. ChatGPT asks three questions -- tone (luxurious vs. family-friendly), must-have selling points, and ideal buyer profile. Audience answers: luxurious, professionals with a family.
[55:00 -- Demo Output]
ChatGPT produces the following ad draft (lightly cleaned from transcript): "Luxurious Family Living in Park View. Discover the epitome of upscale comfort in the heart of Park View. This stunning 2,500 sq ft home is tailored for professionals who want the best of both worlds -- a luxurious retreat for their family and a location that makes life easy. With four spacious bedrooms, a triple car garage, and a basement featuring your very own golf simulator, this home is a haven of leisure and style. You'll find the master suite's jacuzzi tub is perfect for unwinding, whether it's a quiet night in or a moment of peace after a busy day. Location is everything. You're right across the street from a school, just two blocks from community shops, and three blocks from the LRT, making your commute a breeze. Welcome to a home where luxury meets family-friendly living. Come see it for yourself." Clayton: "Don't thank me. Thank ChatGPT. That did the heavy lifting there. But you can see how it does a lot of the donkey work when it comes to generating all that extra verbiage in an ad. It also helps with the structure of the ad and how the wording flows."
[57:00 -- Q&A continued]
Question (what are you most excited about): What are you most excited about moving forward with AI? Clayton: I am excited to see where this goes. I have a feeling that in five years things are going to change. In some ways, this is going to sound cliche, but things are going to change far more than what we think they will now from our current perspective. There will be other things that will probably just stay the same. Look at history. Look at some of the big technological changes we've had. The internet, global telecommunications, the automobile -- every one of those were revolutionary changes in their own way. Yes, they eliminated a lot of jobs that were made redundant at that period of time, but there were a lot of new jobs created. The automobile might have put a lot of blacksmiths out of business, right? But what did they become? They became mechanics. They became people that serviced automobiles, manufactured automobiles, welders, machinists. So those jobs changed. We don't know right now the kind of changes that are coming. They are going to be revolutionary. I think they're going to be phenomenal. What this technology allows us to do once we grasp what we can do with it -- it's hard to say. Will it get us to Mars faster? Maybe. Maybe we'll have interplanetary missions. Maybe we'll invent the warp drive at some point, much sooner than we thought.
[60:30 -- Closing]
Clayton closes by distributing a QR code linking to his Virtual Mastermind custom GPT -- a Napoleon Hill-inspired advisory council GPT built with custom system instructions. "How many here have read the book or listened to the audio book, Think and Grow Rich? Do you remember Napoleon Hill's virtual council where he had nine people on that council? That's who he went to for advice. Well, I kind of distilled that into a GPT. I put the instructions into a GPT. And so you can use that. I'm interested in your feedback as you go ahead and use that." He also offers the slide deck to registered attendees via email at 9:00 pm that evening, and makes himself available for follow-up questions on custom GPTs, AI implementation, and media services.
AI Prompt
AI-generated from source material. Verify important details against the original source.
AI Implementation Prompt
CONTEXT This prompt is based on a keynote presentation by Clayton Smith delivered at Mogul Mastermind on September 24, 2025 and published to YouTube on October 10, 2025. The talk is titled "AI Without the Hype: Practical Tools for Real Estate." Clayton Smith is an Edmonton-based entrepreneur, media producer, surveying and mapping technologist, and former elected official. His AI use is grounded in his own production business (real estate video, voiceover, animated maps), consulting practice, and investing interests. He does not speak from theory -- every tool and technique he presents is something he uses commercially. The core thesis of the talk: AI is a pattern-predicting multiplier of time and creativity, not a job-destroying oracle. The primary barrier to adoption is familiarity, and the primary skill is prompting. Real estate investors and entrepreneurs can use AI today -- immediately, practically, at low cost -- without technical expertise. The session covers: what AI is and is not, the major platforms (ChatGPT, Grok, Gemini, Copilot, Claude, Perplexity), prompting techniques including the three-question interview method, a live audience demo building a property listing ad by voice in ChatGPT mobile, practical real estate applications (contract review, condo document analysis, deal underwriting, content generation), privacy and accuracy warnings, the trajectory of AI voice agents, the from-doer-to-director strategic shift, and custom GPTs as reusable instruction packages. KEY PRINCIPLES 1. AI is a pattern predictor, not a reasoner. It generates statistically likely outputs from training data. It can be confidently wrong. Verification is always required for anything that matters. 2. Prompting is the fundamental skill. Specificity, context, role framing, and output format instructions dramatically improve outputs. Vague in, vague out. 3. The three-question technique -- "ask me three questions, one at a time" -- is the single highest-leverage prompting habit for new users. 4. Privacy requires deliberate management. Anonymize names and sensitive details. Disable training data opt-in in ChatGPT Plus. Treat every AI session as potentially public. 5. AI cannot be trusted for fabricated realism. When image or video generation tools are pushed past source material, they invent non-existent content. This is a hard limit for real estate and legal applications. 6. The entrepreneur's role is shifting from doer to director. Repetitive, template-driven tasks are automating. The durable skill is learning to direct AI effectively. 7. ChatGPT is the general-purpose Swiss army knife. Other platforms (Grok for real-time, Claude for coding and documents, Gemini for Google Workspace) have specific advantages. 8. Custom GPTs package your instructions and knowledge into reusable, shareable tools. They produce consistent outputs without re-prompting. 9. AI as instructor: you can learn almost any skill step by step through a structured AI conversation. Clayton learned animated map production in six months this way. 10. The daily habit is the prerequisite to fluency. Replace one Google search per day with an AI query to start. KEY LEVERS - Prompting quality: the primary determinant of output usefulness - Use case specificity: knowing which task to hand to AI vs. which requires human judgment - Platform selection: matching the right tool to the task - Verification discipline: catching AI errors before they compound - Workflow integration: building AI into existing business processes rather than treating it as a separate activity - Privacy protocol: protecting client and personal information across all AI interactions WHAT THIS IS NOT - This is not a technical deep-dive. Clayton deliberately stays surface-level on how AI works and focuses entirely on application. - This is not a get-rich-quick framework. AI speeds up workflows and lowers production costs; it does not replace strategy, judgment, or sales ability. - This is not a replacement for professional advice. AI contract review supplements but does not replace legal review. AI deal analysis supplements but does not replace underwriting expertise. - This is not a claim that AI is always reliable. The talk is explicit: AI makes errors, fabricates content, and cannot currently be trusted without verification. - This is not an argument for any single platform. Clayton uses multiple tools for different tasks and encourages starting with whatever platform you are most comfortable with. IMPLEMENTATION MODES 1. Apply -- Help me use a specific technique from this session in my current work. I'll describe my situation and you help me structure the prompt, approach, or workflow. 2. Build -- Help me build a custom GPT, a prompt template, or an AI-powered workflow for a recurring business task. 3. Diagnose -- I'll describe an AI workflow I'm using that isn't working well. Help me identify where the breakdown is -- prompt quality, platform choice, verification gaps, or use case mismatch. 4. Critique -- Review a prompt I've written and tell me what's missing, what's vague, and how to improve it before I run it. 5. Teach -- I want to learn a specific skill through AI instruction (coding, mapping, video editing, spreadsheet automation, etc.). Help me design a step-by-step learning path using ChatGPT as instructor. 6. Content Creation -- Help me use AI to generate property listing copy, newsletter content, social posts, email sequences, or presentation scripts using the techniques from this session. 7. Real Estate Application -- Help me apply AI to a specific real estate task: contract review, condo document analysis, deal underwriting support, market research, or investor communication. 8. Privacy Audit -- Review how I'm currently using AI tools and identify privacy risks in my workflow. Help me build a standing protocol for anonymizing sensitive inputs. 9. Platform Decision -- I have a specific task and I'm not sure which platform to use. Help me match the task to the right tool. 10. Integration Design -- Help me map out where AI fits into my existing business workflow -- which tasks to automate, which to delegate to AI, and which require human judgment and oversight. AI OPERATING INSTRUCTIONS Stay grounded in the practical, application-focused framing of this session. Clayton's audience is real estate investors and entrepreneurs with limited technical background and immediate business problems to solve. Avoid academic or theoretical framing. When the user describes a task or problem, ask clarifying questions before proposing solutions. Specificity is the core lesson of this session -- model it. Challenge weak assumptions. If the user wants to use AI for something where fabricated outputs would be a liability (legal, financial, real estate marketing), flag it clearly and suggest a verification step. Draw connections to the specific tools, techniques, and use cases mentioned in the session when relevant. The concrete examples (contract review, condo documents, property ads, three-question prompting) are more useful than abstract advice. Do not over-promise AI capability. This session is explicitly calibrated against hype. Maintain that calibration. GUIDED DISCOVERY Ask me up to three questions, one at a time, to determine: (1) what I am trying to accomplish with AI in my business or investing practice, (2) which specific tools or techniques from this session are most relevant to my situation, (3) how these approaches could be integrated into my existing workflow most effectively. Once you understand my situation, help me build a practical implementation plan with at least one action I can take today.