[00:00:00] Zach Hammer: Welcome back to the Real Estate Growth Hacker Show. On today’s episode, we are gonna be talking about How The More You Understand The Way The Human Brain Works. How you can teach, train, and learn as a human, can actually make a massive difference on how well you understand how to work with AI and do prompt engineering and develop those systems.

[00:00:19] Zach Hammer: And so, we’re gonna talk about that and how that’s kind of a counterintuitive concept compared to the way that most people would think about getting better at working with software historically. So that’s what we’re gonna be talking about today. With me today, I have my guest co-host, Charlie Madison from Referrals While You Sleep Realtor Waiting list working with lenders and realtors and he’s a developer, he’s a founder. He is not wearing Hawaiian shirt today, which is a surprising thing. But we love them just the same. Welcome Charlie.

[00:00:50] Charlie Madison: Hey,it’s good to be here. This is a pretty fascinating topic.

[00:00:55] Zach Hammer: Glad to hear it. So, yeah, so let’s set a little bit of the context here. So, the [00:01:00] reason why I feel like this is even worth talking about is, some of these concepts might actually be easier than people are making them out to be, but mainly because they’ve thought about them in the wrong way in the past or in, in what’s worked previously doesn’t work as well here. Right?

[00:01:15] Zach Hammer: So in the past, if I’m gonna get really good at working with software, like I need to understand technical specifics to also get an edge software wise. I potentially like, depending on the software and depending like, so like take this for what it’s worth, it’s sort of rough ideas. But you know, like I might need to understand things like weather processes can run asynchronously or whether multi-threading needs to happen or technical limitations how much ram I need. Like a specific exact phrasing and context of like if you think about building out formulas in a spreadsheet, right?

[00:01:47] Zach Hammer: Like if you get even one symbol off. Like, it just doesn’t work, right? Like it just breaks completely. It does nothing like you want it to. And so we’re very used to like software being very particular and [00:02:00] very other, right? Like, we’re used to just sort of like turning off completely the creative side of our brain, the human side of our brain, and going almost exclusively into trying to think like a robot. Right?

[00:02:12] Zach Hammer: Does that check out with you? Does that make sense for kinda what it normally feels like to work with software?

[00:02:17] Charlie Madison: Yeah, it reminds me of, you ever seen the videos of the parent? Where the kid tells the parent how to make a peanut butter and jelly sandwich.

[00:02:27] Zach Hammer: Right. Yeah.

[00:02:28] Charlie Madison: And the kid, you know, skips over all sorts of steps, but the parent does exactly what the kid says and nothing happens. And you know, it’s to teach the kid that you need to be very specific.

[00:02:41] Charlie Madison: And that’s really what it’s like to write a program. Like, I have to imagine that the program it is completely ignorant and stupid and has only like, it doesn’t even really understand what we write. It just, but it’s just line by line doing the exact [00:03:00] thing. And that’s the way computers have always been.

[00:03:03] Zach Hammer: Right. And regardless of whether you’re writing code or just working with software like that, like literally, even if you’re playing video games, right? Like you play a video game and if you press the wrong key to try and shoot, like it just won’t shoot. Right?

[00:03:17] Zach Hammer: If you tell it to go forward, but you’ve got your keys mis-aligned. Like it doesn’t do even remotely what you’re trying to do. Right? So it’s like everything that we’re used to in software up till this point is very typically like, you either get it exactly right and get the result that you’re wanting hopefully, or if you get it even a little bit off, it’s just completely nothing.

[00:03:39] Zach Hammer: Anyway. So we’re used to sort of thinking about like working with software is this exact process of like, you need this magical thing in this exact way in order to get the best results. And there might be some smidgen of truth to that, right? Like working with AI, in terms of people have tested lots of different phrasing and sometimes this phrasing seems to more consistently get a result than [00:04:00] this phrasing.

[00:04:00] Zach Hammer: But it’s not like it’s much less pass fail, right? And so that by itself is already better. That’s part of why people really like working with, you know, large language models in terms of getting an answer, ’cause I could type something, have three typos in it and only basically get at the idea.

[00:04:15] Zach Hammer: And what it’s gonna give me back is still in the right ballpark, right? It’s I’m just gonna respond with, man cannot compute, false, no error.

[00:04:26] Charlie Madison: Yeah, cannot

[00:04:27] Zach Hammer: It’s.

[00:04:28] Charlie Madison: words.

[00:04:29] Zach Hammer: Yeah, exactly. It’s like learn to type you caveman. So like we don’t get that result and so that’s part of why that’s been beneficial.

[00:04:39] Zach Hammer: But even like as we start to look at how do we get like, advanced results from AI where we really dial in, like, things like it, very clearly understanding exactly how to give us back what we’re looking for, right? Like how do we set it up for success on that? How do we set it up so that it could potentially tap into knowledge that we [00:05:00] don’t have in order to give us a better result than we could get on our own, right?

[00:05:03] Zach Hammer: Like, how do we set things up like that? And so, at a basic level, this is what we’re seeing, and it’s a surprising thing that makes sense. But at a basic level, here’s the principle. The more you look at working with AI from the vein of how do humans learn and understand and how are they influenced, what psychology matters to humans, especially when it comes to language, right?

[00:05:30] Zach Hammer: Because remember, that’s how these things are trained, they’re trained on the relationship between language and whatever we’re wanting to do. Even as of right now, even the visual stuff, it’s still all trained on language, it’s trained on how images relate to the words that we use to describe them. Right?

[00:05:47] Zach Hammer: And so ultimately video would be the same thing. How video relates to the words that we describe them, right? How music relates to the words that we use to describe it. And so it all comes [00:06:00] back to language. So when you think about like the psychology of language, when you think about the way that we use language to influence, to learn, to communicate effectively.

[00:06:08] Zach Hammer: All of those things are coming into play with AI. So, like, here’s a, a couple of examples. So one that I really like, you can get a very meaningful difference in the quality of your response if you tell ChatGPT to explain its reasoning before giving you the answer. So literally just have it like, think out loud is the idea, right? Like the same way. The way that we think. If you sort of think of like we come to our conclusion by building our final answer on the scaffolding of reason. If you actually like show your work, write out your reasoning, think it out in advance and do that preliminary step of thinking about something in advance, you’re probably also going to get a better answer on the other end when you’ve had the chance to sort of sit with [00:07:00] the thought for a second. Right?

[00:07:02] Charlie Madison: Right.

[00:07:02] Zach Hammer: And AI does the same thing and it’s amazing that’s a thing that works, that you could literally explain your reasoning before providing your answer, the answer will be better most of the time than it would’ve been if you didn’t ask it to explain its reasoning. Another one that’s similar but different is asking it to think step by step. Right?

[00:07:25] Zach Hammer: So thinking, so same way, right? Like, uh, you were just describing the process that, you know, kind of the funny thing that people do with the peanut butter. But like, that’s a really good example of when somebody thinks through something very well step by step, it also deals with a lot of the potential problems that somebody might run into, right?

[00:07:42] Zach Hammer: So like, you know, the peanut butter example, it’s like the typical one that a kid starts with is like, all right, get the peanut butter, put the peanut butter on the bread and then fold the bread. And now you got a sandwich. And so like, you’ll see like, they grab the jar of peanut butter, they put the jar onto the loaf of bread, they fold the loaf of [00:08:00] bread and then hand them that, right?

[00:08:01] Zach Hammer: Like, so, so it’s like, okay. I didn’t think step by step up. Okay. So. I’m gonna fix this now, right? So I grab the jar, you’re gonna open the jar, you’re gonna get the peanut butter out, you’re gonna take out two pieces of bread, you’re gonna put the peanut butter onto the pieces of bread, and then you’re gonna close it.

[00:08:19] Zach Hammer: Now I fixed it, and so then the parent does this. They get the jar, they open it, they stick their hand in, grab some peanut butter, right? And so, as you’re thinking through something step-by-step. The more step-by-step, you think about it, typically the clearer you get to the end result. Right?

[00:08:34] Zach Hammer: And so similarly, even just telling it to think through something, step-by-step, even if it doesn’t do that work of like showing you that work in advance. It creates this better result where it’s like, better reasoned and it makes more sense. So yeah. At a basic level, does that concept make sense to you for kind of a way to think through working with AI to get better results?

[00:08:59] Zach Hammer: That’s maybe [00:09:00] less than intuitive for what you might expect for typical software.

[00:09:03] Charlie Madison: Yeah, the fact that, I mean, Google, you know, if you Google something or you open Microsoft Word or nowhere else in computer that I know of, has that ever happened? You know, you don’t go to YouTube and say, show me all of the, you know, videos on fire hydrants. I don’t know why I said fire hydrants. But please explain your reasoning on why you chose these, Right?

[00:09:27] Charlie Madison: It wouldn’t make a difference, like it would actually hurt the results, but here it’s a brand new world where it gets it to, I guess, clarify what it’s doing.

[00:09:36] Zach Hammer: Right, yeah. And like the reasons why are interesting, the reasons why this works may not be the same reasons why it matters for us. It may not be the same reasons why it helps us to think better, clearer, etc. But what it does is essentially these large language models at a very basic level, they’ve been shown a lot of text and they’ve been used, you [00:10:00] know, they’ve used that in order to learn how to predict, when I’ve got this set of text, what is the most likely text to come next.

[00:10:08] Zach Hammer: And they get better and better at predicting the text that’s gonna come next. And so you could potentially imagine that in the context of, okay, I was shown the text, think step-by-step. Explain your work, explain your reasoning, and you compare the kind of answers that come after that. They might typically be your better answers to the ones that didn’t take step-by-step that didn’t explain the reasoning, etc..

[00:10:28] Zach Hammer: So, like that could be part of the reason why you end up getting to a better, clearer. You know, more detailed, more thought out, better structured, whatever. But what’s interesting is if it helps humans get a better result, it probably helps AI even if the reason why is completely different,

[00:10:45] Charlie Madison: Right.

[00:10:46] Zach Hammer: Right? And so here’s kind of the key takeaway from this. If you want to get better at foundational principles of working with these large language models and getting a result that’s consistent, that’s at a higher quality, that’s at a higher caliber [00:11:00] than maybe what the average person does. You don’t need to be like just looking for the next trick from prompt engineers, right? Like you could get to the underlying principle of what’s helping people think through these things, and now don’t get me wrong. Those can be great shortcuts, they can be great ways to potentially get an edge quicker, but like they aren’t tapping into some mystical science that you can’t learn and understand as well.

[00:11:27] Zach Hammer: And the basic things that are working is if you study and understand things like communication, clear communication, persuasion, influence how humans learn. Like, if you understand those kinds of things and you study those and you learn those, and then try and deploy those principles into how you’re working with AI, chances are you’re gonna start seeing that get you a better result in the end results as well.

[00:11:56] Zach Hammer: And again, the reasons aren’t necessarily because the AI thinks the way [00:12:00] that we do. It’s almost like, I don’t even know how to describe this concept. It’s almost like a mirror, but like a mirror that learns how to reflect back to us based on the things that we put into it, even if the mirror doesn’t work by being exactly what we did.

[00:12:16] Zach Hammer: Does that make sense? Like, if we think of a mirror as being literally just a dumb, it’s light reflecting off of a surface back into our eyes that mimics what, like what we’re seeing, right? AI is almost like a mirror that’s more like a webcam hooked up to a screen. Right?

[00:12:31] Zach Hammer: Where it’s very similar, but it’s like being processed and it’s changing pixels in order to create the result. It’s not the direct thing, it’s going through this digital thing, but in the same way, like if I move my hand here, it still shows up through the webcam on the screen. Right?

[00:12:46] Zach Hammer: Just like it would on a mirror. So the principles are different, but it works the same in the end. And the other benefit to this is that not only does this help you with working with large language models. It’s starting to figure out your own prompts and your own ways of working [00:13:00] with AI that get you the results that you’re looking for, but you’re gonna need the information anyway, right?

[00:13:05] Zach Hammer: Because you’re gonna need to know how to teach people, how to persuade, how to influence, how to communicate clearly, you’re gonna need all that information. And ultimately get to the end results as a human, whether that’s working directly with people or whether that’s working with AI and seeing is it actually doing a good job of communicating.

[00:13:25] Zach Hammer: Is it actually doing a good job of persuading? Having the expertise and the knowledge to say, is this just giving me what I asked for? In terms of what something that it looks like, what I asked for? Is it actually giving me the things that I’ve seen or proven, tested and actually work? Does that all make sense?

[00:13:38] Charlie Madison: Yeah. That does make sense.

[00:13:40] Zach Hammer: What’s a concept from this that you think that you could take and run with and put into practice to test with your AI prompt engineering?

[00:13:49] Charlie Madison: You know, I mean, the simple one is just tell it to think through the steps, right? Like, that’s just the simplest one. Whatever I do and think through the steps that’s simple. The other is the [00:14:00] excitement of wow, like as I’m learning persuasion, I’m learning AI.

[00:14:08] Charlie Madison: As I’m learning AI, I’m like, it’s bringing how to be a productive, successful human. Like, together, and kind of what I’ve been thinking about, like, man, if I had a worker. That knew everything.

[00:14:24] Zach Hammer: Right,

[00:14:24] Charlie Madison: And my only job as their leader was to persuade them and motivate them. What could we get done?

[00:14:33] Charlie Madison: And so I like that kind of avatar of like, this is what I need to do.

[00:14:38] Zach Hammer: Right. Yeah. And what’s interesting about this too, just to go down this rabbit hole a little bit. So there’s a couple of other like, sort of fascinating things that they’ve discovered in this vein. One of them that they’ve noticed a tendency around, I haven’t checked this, so I don’t know how much stock to put into this, but they have noticed that if the large language model thinks it’s winter.[00:15:00]

[00:15:00] Zach Hammer: It will be lazier in its responses than if it thinks it’s summer,

[00:15:04] Charlie Madison: Wow.

[00:15:06] Zach Hammer: Because for a lot of us humans, around winter, around like the holidays, people start slacking off and kind of going into like a holiday mode for a period of time. So maybe you don’t put as much work into your writing. You don’t write as long, you’re still doing what you need to, but you’re not like, being as in depth or thorough about it.

[00:15:24] Zach Hammer: And even though the AI hasn’t been trade, like when it’s winter, you write, you know, you put less work into things. It’s like, oh, it’s winter. Okay, well that means, you know, ’cause it’s like monkey see, monkey do, apparently when it’s winter, this is a factor that means we don’t write as much.

[00:15:40] Zach Hammer: Like that’s the right way to do this.

[00:15:41] Charlie Madison: I mean, it’s really learning like children do, right?

[00:15:44] Zach Hammer: Exactly.

[00:15:45] Charlie Madison: Children, watch what you do and just duplicate it.

[00:15:50] Zach Hammer: Exactly. So like, there are literally things like that. Like, we are still for anybody who thinks we figured all this out and like there’s people that really understand this deeply and know [00:16:00] everything.

[00:16:00] Zach Hammer: It’s like even this is what I’m seeing and everything that I’m hearing from the people who are the smartest people in the room on AI, the thing that I hear from them most consistently is like, we don’t know what this is capable of. Like, we are the ones who built it and understand how it was built, and we don’t actually fully know what it’s capable of.

[00:16:19] Zach Hammer: We’re still discovering new things every day. Right. And that’s simultaneously terrifying and exciting. It’s terrifying because we don’t know this beast that we’ve built and what’s even possible off it, but it’s exciting in the sense of there’s still plenty of opportunity for you and I to discover amazing new ways to work with this system, right?

[00:16:43] Zach Hammer: To throw a level of creativity into it that somebody hasn’t considered before. Like this is very much in the wild west you know, virgin jungle kind of thing, where it’s like literally you might be the first person to think of a concept and to try it out.

[00:16:58] Zach Hammer: And to be able to get [00:17:00] meaningful results from it. You don’t have to just tread down paths that other people have gone down. You can test new ideas and meaningfully expect that it might be an idea that somebody hasn’t tested yet. Right.

[00:17:11] Zach Hammer: And I don’t know about you, but that’s exciting ’cause all throughout life, it’s like growing up, it’s like school is mostly learning about all of the things that everybody else has already figured out and how to do what they’ve already figured out. Right. And even like you get into business. You get into your job, you get into a career, it’s mostly like somebody else has figured this thing out and it’s your job just to implement it.

[00:17:33] Zach Hammer: Or like when you’re starting in a business. It’s like, all right, first I gotta wrap my head around what has already worked up till this point, then I might be able to start figuring out some new things. So it’s like we spend most of our life literally just kinda repeating the same things that other people have already figured out.

[00:17:48] Zach Hammer: It’s really cool that there’s a major segment that’s this wide open frontier for any of us to explore that we’re kind of on this maybe not equal playing field, but at least a lot more [00:18:00] equal than it was before. And when you discover things in that frontier, they can make a drastic, meaningful difference in your business, in your life, in your work.

[00:18:09] Zach Hammer: And yeah and the tour, like the tools to do it. Are the things that we’ve already discovered that are just slightly differently deployed in this realm and like how they make a difference.

[00:18:19] Zach Hammer: I don’t know about you, but I find that really exciting, really interesting and you know, I’m excited to be part of going down that ballpark. What about you?

[00:18:26] Charlie Madison: Yeah, it’s an amazing time to be alive. I think, it’s a time where you can be a craftsman, going through the Dan Sullivan Strategic Coach. Like if I could boil it down to anything, it’s successful entrepreneurs usually got there by being an amazing lone wolf.

[00:18:43] Charlie Madison: And the first three years of Strategic Coach is teaching you how to find out what you’re really good at, he calls it your unique ability

[00:18:52] Zach Hammer: Right.

[00:18:52] Charlie Madison: And then be able to delegate and delete other stuff. But then the next three years is [00:19:00] all how do you build a team of people? Who have unique abilities on other things required in your business.

[00:19:06] Charlie Madison: So no one’s just halfway doing it, but you’re actually leading a team of, I call it people in their superpowers. And you know what I think AI is going to let us do is we’re going to be able to have many more teammates through AI and the human teammates we have are gonna be like Iron Man, you know, they’re gonna be exponential because they’re going to have AI kinda like, I think his name’s Brian Roemmele says, he calls it a IA, your Intelligence Accelerator.

[00:19:40] Charlie Madison: So like an individual can just create brand new things that a much more rapid pace. And from what I can tell, every time that there’s been a major invention, it’s made the world better. And it’s created more jobs and so that’s really cool.

[00:19:58] Zach Hammer: Yeah, [00:20:00] absolutely. So there you go. So, to sort of put a bow on it again the short concept here. Is that if you wanna get better at working with these large language models, if you flip the script and instead of thinking there’s an exact right way to do this that somebody else has already figured out, and I just have to learn that and implement it.

[00:20:19] Zach Hammer: If you switch your thinking from believing that’s the case to understanding, sure, there’s lots of opportunity to grow and learn in this thing together. But there’s also a ton that hasn’t been figured out and further, the ways that seem to be working are by better understanding humans and the way that we work, you’re able to find those tools that seem to work well with AI as well, and go down that path in order to discover those opportunities.

[00:20:46] Zach Hammer: So, start deploying that and further, come back to us. Let us know what you are testing because just like Charlie was just mentioning, we each have our own unique abilities. We are each going to be going down these things and thinking about them in different ways.

[00:20:59] Zach Hammer: In [00:21:00] your own unique framework around what problems are you facing day by day, what things do you enjoy solving? What rabbit trails do you enjoy going down? And you’re gonna go down a path that I’m not gonna go down. And you’re gonna go down that path and discover interesting things that’ll be useful for me because you are uniquely you, and together we can learn those things together at a more effective rate.

[00:21:20] Zach Hammer: And so if you’re interested in doing that kind of thing, where not only are you looking at putting together the work of AI and where it is making a difference in the real estate business into practice and actually getting down and putting into work together.

[00:21:36] Zach Hammer: But if you’re interested in coming together with a group of like-minded people who are all doing the same thing as well, that’s something that we’re working on putting together right now with Real Estate Growth Hackers, where we’ve got an AI mastermind, an AI group coaching sort of thing.

[00:21:48] Zach Hammer: I still don’t have the right word for it, but what we’re doing is we’re getting together and for high-level teams who are putting this stuff into practice in marketing, in recruiting, [00:22:00] in back office, right? In all of these areas of the work of real estate.

[00:22:05] Zach Hammer: If you’re interested in coming together, putting this stuff into practice, seeing not only what I’ve learned and figured out, but what other teams across the country, across the world, probably at some point are figuring it out as well. I’d love to invite you to come and join us at that.

[00:22:18] Zach Hammer: If you’d like more information about that, you can go to RealEstateGrowthHackers.com/Contact. Let us know that you’re interested and we’ll get you the details on how you might be able to join up with us. It is not the kind of thing that we’re going to be bringing a ton of people into.

[00:22:31] Zach Hammer: We got at least one or two right now that are already coming in and ready to go. I am literally thinking we might stop it at about 10 before I consider pulling in more people. I don’t know how big it makes sense to let it get before it starts having negative ramifications for everybody involved in terms of not being able to learn together effectively.

[00:22:52] Zach Hammer: Just we’re, it sort of breaks the dynamic of it. So, you know, I don’t make any guarantees about how small it gets, but we’re gonna figure it out as it goes. And 10 to me is kind of that [00:23:00] number that I could think of. I feel like we could safely get that big and have it still work, like I’m envisioning it.

[00:23:05] Zach Hammer: But yeah, so if you’re interested in that, I recommend not delaying, first come, first serve. We’ll be getting people in and getting people involved. RealEstateGrowtHackers.com/Contact.

[00:23:16] Zach Hammer: Any thoughts or questions on that Charlie that you think I should address?

[00:23:19] Charlie Madison: It makes sense to do it. Get in there, get involved with other realtors. It’s amazing what all you guys will be able to create when other people are still scratching the surface.

[00:23:29] Zach Hammer: Yeah, I mean, I honestly I’m half convinced that the main reason why Charlie keeps doing this show with me is just because he gets unique access to being able to get some of this information every week. So, if you want to have that same kind of ability to have me in your back pocket and do the same thing with other real estate agents across the country that definitely check out what we’re up to RealEstateGrowthHackers.com/Contact.

[00:23:52] Zach Hammer: And then I would be remiss if I did not mention Charlie Madison again, Realtor Waiting List Referrals While You Sleep. If you are a [00:24:00] lender looking to build a waiting list of realtors, rather than having to hunt them down yourself actually have them hunting you and wanting to work with you.

[00:24:10] Zach Hammer: I definitely recommend checking out what he’s up to. We have a link set up RealEstateGrowthHackers.com/ ReferralsWhileYouSleep. I think we will be setting up RW. What is it? YS? I know. Yeah. I tried to spell out loud while I’m. Visualizing it is hard. RWYS, we’ll make sure to get that set up, I’ll do that right after this. So that’s live as well.

[00:24:31] Zach Hammer: You can check him out there. Charlie, what will people see when they go and check you out? What you’re up to when they go and visit that link?

[00:24:37] Charlie Madison: They will see the special VIP private calendar link for friends of Zach and, my most open calendar. So find some time that works. No sales at all, just find out what you’re doing, find out if there’s a way that we can help you and that’s it. Make it real simple. You’re a friend of Zach’s, so you’re a [00:25:00] VIP.

[00:25:01] Zach Hammer: Perfect. I definitely recommend that you guys check that out. In terms of a system to market your business and stay top of mind, whether you’re a lender or a realtor. It is really, one of my favorite ways right now that I think makes sense for everybody to deploy and implement.

[00:25:14] Zach Hammer: No matter what you’re doing, it will make what you’re up to better. ’cause it will make sure that you have more awareness in the easiest ways possible, which is just as important. Not only will it get you in front of people more effectively, it makes it easy for you, and that makes a world of difference for actually getting the results.

[00:25:30] Zach Hammer: So, definitely check out what he is up to, realEstateGrowthHackers.com/RWYS.

[00:25:37] Charlie Madison: Nailed it.

[00:25:39] Zach Hammer: Ah perfect. All right. And again, if you’re interested in our AI Mastermind, check us out realestategrowthhackers.com/Contact.

[00:25:47] Zach Hammer: Charlie, thanks so much for coming on again. Hopefully we brought some people some valuable insights and working with AI today.

[00:25:53] Charlie Madison: I love it. Thank you so much.

[00:25:55] Zach Hammer: All right, bye everyone.