🛜 Connect AI to Your Email, Calendar and Deals (Watch It Go to Work) with Guillaume Black

Hello team and friends, hope you're doing really well. Great to see you guys. So let me go, let's go and share my screen and we're going to get started straight away. Right, so I got the right one. Oh yeah, cool, share my screen. Okay, so we are continuing this, um, 4-part series over the— this summer, you know, we're doing this, um, Thursday AI Takeover. And the idea is really to give you some context, knowledge, and really expertise on the journey we've been on ourselves over the last couple of years, how we've transformed really our operational business, and now helping others and some of you guys already do the same. So AI is really, really powerful, but it can also be very distracting. If you just play with it and don't use it in ways that are going to make you money. So the last couple of sessions we talked about basically the 3 things that matters, you know, in terms of making AI working for you and why is it that it gives you average responses and things like that. So in this part we're going to talk in more details about your context itself, or it's where it sits and the problems when you DIY it. And I want to explain you some of the challenges we solved with that and how basically you can have something that compounds with you all the time of your own, like really use AI natively and straight to the models. without having to rely on lots of other tools. Okay, so, right, so welcome everyone on today's Thursday's AI Takeover. And as I said before in the previous videos, if you've not watched, there's 2 recordings and 2 videos where we talked about, yeah, how do you make the most of AI. And today, as we follow through in the same series, I want to show you exactly tactically what I mean by context and where it actually lives and why it's the only leverage and lever you can have to use AI properly and not being— to be an implementer of AI, not really being a user and someone spending money on lots of tools that don't make sense. So where is it that your context actually— So before we start, just a reminder, I've not come out of the woodworks to figure out how AI works. We've been creating AI solutions for property investors, you know, for 7 years now. And, you know, obviously, you know, Propertyfilter. And just in the last 12 months alone, we've seen more than 3,250 offers accepted on the community, a community of 1,800 of the savviest and most active investors in the country. And that's a quick overview so you know that It's not coming out of nowhere. So 3 things when it comes to this context and how important it's going to be. It's really about cloning yourself. It's really about an organized way of your AI to know you intimately, deeply, and know you really, really well. And the idea is to turn everything that comes into you into a context, into proprietary data, and actually makes your AI second brain work really well. Your property brain or whatever you want to call it. And this is really important. So it's like having an employee and you can't expect an employee to do his job if you don't train them, if you don't explain them. But the benefits with AI is you train them once and they never forget. So 6 months later, it will pick up stuff you've told them back in March or whatever. And if any one of you have had to train VAs, I've worked with team members, I've coached people, you know the frustration of repeating things many times. People not taking notes and things like that. And the really attractive things about having AI to do things with you is it never forgets you, which means that you can compound exponentially the things you can do because whatever you do, you build on top of it every time. So that's really the first thing. And now you can turn basically all sorts of a pile of files, I call it, all things to do with what we do into an organized brain with connections and things so that it's really going to work. And the philosophy here, and the reason why, and the bet I'm taking is that I bet I can help you understand how to use AI natively, straight to the model, without paying the ignorance tax of basically subscribing to all sorts of tools that you don't need, because all you need is basically the context. And we're going to talk about this as well, how this folder, if you set up properly, will outlive the likes of OpenAI. It will outlive the likes of Anthropic and companies like that. Because it's, you know, this will be the most valuable folder for the rest of your working life and the rest of your life if you expand into doing personal things with it. So I said before, you know, the success of your implementation with AI has got nothing to do with the tool. You know, it's not about the overwhelm about which tool do we buy or which AI model we use. It's really about the context you own, this proprietary data, which is really— this context got 2 parts. One is the niche knowledge, niche data about your industry. We know what's actually current, the strategies working right now, or do you negotiate with vendors in today's market? Really everything that's topical and working right now, not the average of the training data of the internet since 2020, 2024, of stuff that don't work right now. If you ask your AI about property questions, its training data are out of date. So it will give you a very confident answer. It will tell you about strategies. None of this shit works, right? So this is the reality. So really the value is in you onto building and maintaining a context, which is 2 parts. One is expertise in your niche of stuff that are currently working right now. And 2, the data about you. So everything to do with your strategy, your constraints, how do you basically apply the reality of the world right now to your actual context, the constraints, the time you have, the finance you have, the access to the network you have, the priority you have, et cetera, to actually get to the outcome. So that's what I mean here by context is your niche data and your own data, as it were, that you and all of this you get to own. So none of this will exist inside a chatbot. You don't open a tab, go into ChatGPT and expect it to run your business for you. All the hype you see on Twitter, all these people who are having businesses that run on AI, you can imagine that they are not using ChatGPT like it's 2022 again. So they're not opening a tab, copying, pasting things in and out. No. They own their context and they use AI in the way I'm sharing it with you. And the mistake we've done when we— so basically the big aha moment we had months ago now is the time where we threw— there's different AI interfaces you can use. And then within this interface, you can call a model. We talked about this in the last video. And you connect your AI interface with your tools, your emails and things like that. Calendar, Drive, and then a lot of other things, whatever you want, basically. And you can point your AI interface to a file on your machine, and this file becomes your context. And this is really the magic of all of this is because you don't need to become a— you don't need to be like a prompt engineer, like the very long prompts, like you are a marketing expert, you are a property investor, like the 7-part prompt structure to basically cater for the lack of context the AI, the ChatGPT browser has on you. You don't need any of that because it's pointed at your context and your context can be organized as a brain. So I'm going to tell you a little anecdote. I don't know if it's kind of covered here. So have you ever had the case where— and everyone's had this case where you receive an email and there's an attachment and you want to you want to go and save this attachment on your filing system, on your drive, or on your computer, or whatever. And then you ask yourself, this file could go in 2 different places. So this is to do with finance, but should I put this in the company, like sort of finance topic, or should I go inside the project of that property and I store it inside the finance aspect of that project? So decisions like that, you probably take 1 or 2 a day. That's quite normal in your business, or even more. But the thing is, when you do things with AI, it can take 1,000 of these decisions a minute. So it's really, really dangerous when you point your AI at a folder without a structure. It will go and create a mess. So if there's more than 2 of you working on the business and you don't have a right convention about where things go, things are going to go all over the place. You've seen this at work on shared drives, on all sorts. Yeah, it's a mess because there's no agreed convention and everyone does whatever they want and things are all over the place. It's a bit the same with AI if you let it loose. So you just need to have a convention that works. And to help you, what we figured out as well is AI works by utility. When you ask it a question, it wants to go to an outcome. It's built to go and solve a problem. But the way a lot of our traditional filing has been working is we file by topic, we file by marketing, we file by maybe projects over here. And that's our tradition we've done. But AI is a lot more efficient when you instruct him to file your context and organize your brain by utility. So because you're going, okay, what problem am I trying to solve? And then the whole architecture comes down from that. That's the first thing I want to say. So one is, we don't— your brain, if you go and set up your brain yourself, or if you go and create your filing system to maintain, to create and maintain your own context, just be mindful, it will create a mess if you let him loose. And secondly, when you start to get organized, we bring our own bias into this, and we want to create a topical, organization, it's not optimal for AI. It's rather— it's much better at basically following utility. So I'm going to share with you the exact file structure so you can use it yourself. And also you want to make sure that this whole organization is mutually exclusive and exhaustive, which means that there's a place for everything and everything's got a unique place. So the way your AI is going to be guided, there's no questions, you know, it's quite binary. It's black or white. This file can only go in one place. And so that's again, because of the way to do that. I'm going to come back to this in a second. But basically, if you look at the property AI brain, you know, like we are offering to our property filter members, there's 6 drawers, 6 parts of that filing system. The inbox, project, flywheel, knowledge, and then you've got the tools, and the archive, which I'm not going to cover today. But, but your 2 main folders are going to be your projects and your flywheels. So the flywheels are the things that never end, the things that are always true about you. A flywheel could be, um, so for example, in property, there will be like your deal flow will be flywheel. So you're always sourcing deals, you know, like all your, your machine that source deals is always working, you know. So If you don't, if you think finding deals is a project, you start, stop, start, stop. You know the whole philosophy around that. You are in the business of building momentum. If you are not looking for deals all the time and you know the blueprint, right? It's as little as one viewing a week, one offer a week. Property gets really, really hard. It's really, really hard if you go and do 20 viewings a week for 3 weeks. and then nothing for 2 months, uh, it's really, really hard. It's really hard if you join PropertyFilter, you get a deal within 3 weeks, and then you stop, and then you come back, you know, 6 months later, 9 months later. You cannot build momentum. You have to understand the rules of the game and play to win. So the whole finding deal blueprint is a flywheel, you know. So we are always sourcing. It's as little as one deal, one offer a week, you know, one viewing, one offer a week, so that You build momentum so that you are seen active in the market, so that you are just with one offer a week. You know, that's 50 offers a year. And then, you know, the end of the blueprint, you know, we follow up on everything that we do. Within 6 months, we are following up on 25 properties, almost 25 offers. So every agent in your area hears about you every week, you know. And that's how then they think of you as the go-to property problem solver. They bring you deals. They only want to sell with you. They think you're the best possible buyer, et cetera, et cetera. So that's the blueprint. So your deal flow is one, is one of the flywheels. Your portfolio, and you might have multiple portfolios, you know, you have multiple companies, interest in different companies. Maybe you have JVs where you've got 50% of certain companies. Maybe you have JVs where you've got 20% of other companies. Every business interest, you know, like every share you've got in something, and maybe you do stocks and shares separately, all of these would be different flywheels. You know, there's no limit of the amount of flywheels, or if you think of it like hats, you know. And then as you expand your use of the brain, you'll see, you know, it does a lot of the life admin. And so for example, me, my own brain, there's obviously like my different property portfolios, there is the property filter, there is my role inside property filter, there is my role as an investor. And I've got my personal stuff, you know, like I'm a parent, I'm a dad, you know, there's I'm like a family member, and there's things I want to do for my groups of friends that are flywheels, that are always true about me. The projects are things that have got an end, which is why I've got this slide here, because this is basically the main fork, you know, the main fork inside this brain architecture. There's, you know, there's a lot more to this, but basically the projects are the things that have got a start and an end. This is really really, really clear for the AI because it's really— you can see the utility. So when you are working on the admin of a refinance, for example, although this is part of a flywheel, part of a portfolio you've got inside your flywheels, it's a project. It's got a start and an end. The admin starts when you instruct your broker to look for a refinance. Then you have all the admin of filling forms and providing information and 70 back and forth emails for your refinance. And then it ends with completion. And then at the end of that, it's archived. So it moved out of project, moves into drawer number 6, the archive. Okay. And so if you think of a refurb, it's the same, et cetera. So anything that's got a start and end. And it's really, this was a game changer in the way we managed to compound the utility of AI for us because it was really, really clear about putting everything in the right place. as I said before, mutually exclusive and exhaustive. So everything's got— there's a place for everything. Everything is in one place. So it's really, really well organized for the AI to really go and do lots of things. Other projects. So you can think, say, refurbs, obviously. So refinance, like the sale of a property is one. Once you've had an offer accepted, the conveyancing is one. The deal progression is something you actively want to be doing. All the deal sources are there. I think it's hard to do deals. Let me tell you, there are deals everywhere once you follow the blueprint on PropFilter. But then getting the offer accepted is the easy part. Then you need to project manage to be on top of everything, the deal progression. And it's really where people drop the ball and struggle to scale their sourcing activity because they are choking under the amount of admin that is required to be on top of, uh, you know, deal progress, the deal progression basically. And these are projects, you know. And because it's all of this happens by emails and, and stuff like that, your AI does all your deal progression, your AI does all the admin and the, you know, the chasing of the solicitors, the brokers, uh, when you, uh, when you do your purchase, your sale, your refinance. All of these are projects you're managing there. So these are flywheels and projects. And the knowledge is— so that's really interesting. So this is really where AI lets you down because the knowledge available out there is its training data and what's on the internet. So you'll have— it's good at researching average stuff, give you average response, but you get the same as everyone else. Your edge, you know, that's why, you know, some of you might, might be going on courses. That's why some of you might be getting coaching. That's why some of you are regulars to, uh, all of the, you know, like the 8 weekly live calls we run inside the Property Filter Community. You know, you build knowledge, you know, you take notes. So now how can you create a system so that all of the courses you've bought, you know, you, you basically feed it inside your AI and they are stored inside your private knowledge. Inside there. So there's 3 layers. I don't know if I've got it in there, but there's 3 layers of your knowledge in the way the brain is organized. And I suggest, you know, either you can basically get it from us or you can build it yourself. But think of it where your competition using the available AI, it's going to the training data and the internet first. And that's the average, the beige. And so you want to flip this on its head so that it goes first to your proprietary things, the way you do your business, all the things you've learned, all the courses, all the things you want to feed into it. First it goes to your knowledge. Level 2 is can you tap into proprietary knowledge bases and secrets that are current with your industry right now that are not publicly available on the internet? So that's, um, so for example, in PropertyFilter, we have, uh, we have created the MCP and the API into, uh, the property, you know, into the PropertyFilter ecosystem. You access all the motivated sellers, you know, from the software lead generation software. You access your lead generators, your pipeline, and all of these things, as well as all the knowledge base, you know, like, because, you know, with 1,800 members doing deals every day There's more than 1,000— there's been more than 1,000 hours of Q&A, live coaching inside the community, you know, in the last couple of years. So there is no question that's never been answered before that the AI doesn't know, you know. So this is why, like, for example, like working with us, level 1, you get obviously like all the proprietary, all the courses, all the stuff you want to teach it, you give it to it, all the PDFs you've ever downloaded, you know, that's your your private knowledge if you want, the books, you know, you've read and the, the things that resonate with you. Level 2 is the proprietary, um, knowledge base you can, uh, tap into, you know, if you've got access to. So for example, this is like the most current things about property you will get by the— with the PropFilter MCP. And only after that, level 3, you go, you go and interrogate the internet to see, uh, what's the average out there, you know. And so the world has changed. If you buy courses right now, it's not just for you to put it in folders and take notes, put it in folders and stick it in a cabinet somewhere. Everything needs to compound. So imagine you have a personal AI system that sits with you and never forgets anything. So you can give it all these materials and et cetera, and you paid for it. So it's for you to use, in my view. And then the inbox, so it's first, but I finish with it. The inbox, what we call the inbox in the filing system, is your ability to, because your knowledge here is going to be like static knowledge. When you do the onboarding on the Property AI brain we provide, it's going to ask you lots of questions so it knows you, so it knows your business, creates all your flywheels, all your projects, and et cetera. And, and, but that's static, you know. Or is your AI aligned with the activities you are doing today? You know, how does it know the actions to be taken from the different meetings, from the different meetings, and, and etc.? So the inbox is a way for you to connect tools like Granola. You know, Granola, there's lots and lots of others, but we do recommend Granola inside the AI brain. It's something, if you think about it, it used to be like Fireflies and all sorts of really annoying bots that sit on Zoom calls, et cetera, that really you break rapport with anyone. It's just not great. The Granola, and there's others, I use the Jamie, which is a German one, European one, Jamie AI, they're all the same. Basically, they sit on your phone or on your laptop. You turn it on, it sits outside of every Zoom call. Any conversation, you know, like any Zoom meeting. Like, it's not a bot that sits in a meeting, and it transcribes all the Zoom meetings you have. The in-person— you can use it for in-person meetings, you can use it for taking ad hoc notes. So I've just been at the viewings, this is the refurb, you know, we're gonna need new carpets, new kitchen, uh, new this, new that. We're gonna— the windows at the front are fine, we need to replace all the ones at the bottom. Every note you capture, capture, capture in there. And then you create commands and automation, routines inside your brain so that it trickles this down to your projects, your flywheels, the knowledge. The knowledge is captured, the flywheels are captured. For example, in your flywheels you'll have all your to-do lists, et cetera. So I am totally paperless. I only sit in meetings, my AI takes notes and adds to my to-do list inside my flywheel. And anything to do with a specific project is automatically trickled down like that. So that's where we've organized all of that. And then if I go into more details inside the flywheel, so that's basically the all, if you like, utility-based folder your AI wants to point at. I'm just sharing with you exactly all of it, right? And then the other 2, one is Number 5 is the tools. So this is where all your commands and skills sit. So a skill to negotiate, a skill to role-play, a skill to profile vendors, a skill to write offers that get accepted, a skill to create comparable reports, a skill to— any tool you think you need in property that produces anything, you can have it natively now within your AI interface. It creates all of these things. If you want to deal a deal pack to sell your deal, it's a skill in your in your tool, the folder number 5. And then folder number 6 is just your archive. When your projects are done, they get into the archive basically. When your meeting transcripts are processed and trickled down into the project and the flywheels, they are archived as well. So now if we zoom in the flywheel, the 7 flywheels, I'm just going to share with you an example of 4. So I kind of talked about this earlier. The things that are always true, always true about you. So the— your deal flow is what I said, you know, it's all the ongoing viewings. And if you think— if you don't have deal flow as a flywheel, as a thing that is your identity, that is always happening, uh, you know, question why you think property is hard, you know, or do something else, you know. So this is really like so basic, like it's like rule number 1 of the Property Filter Blueprint is, uh, You are in the business of building momentum. You know, as I said before, it's the little and often. You know, it's one viewing a week, one offer a week. That's 50 offers a year. You are a top 1% investor. All you need—that's that's it. You know, and honestly, if you don't, and and you're wondering why property feels hard, you know, like this is this is the reason. Portfolio portfolios. You know, so you have maybe a portfolio with a JV partner over here, portfolio with other people over there. You have a a trading business over there in property that buys and sells properties. That's always true about you. In there will sit your sort of processes about things like that. Your power team, your solicitors is always the same. Your broker is always the same. This is important because when you want your AI to draft, create offers, follow up on people, it needs to know who are the people you work with. And again, you cannot expect people to be in your— an employee to guess what's in your head. So we tell it and we literally tell it, none of this is typed. You don't go and type any of that, especially if you get the property AI brain that we have. You just sit through the onboarding. It asks you questions. You literally speak to it. So the world has changed really quite a lot. Your AI interface is an ability for you to talk to your AI or chat, talk to your AI basically. That's it. Wealth is another one. So wants to know about you, your goals. How do we— because if you— for someone to coach you, for an AI to basically be your best mentor, your best personal assistant, and do all sorts of things with you, needs to know the direction. Because success in property looks quite differently for different people. And so you need to be clear. If you are not clear, how can you get other people to come with you? on that journey? If you are not clear, how can you get automation and AI to basically help you achieve these goals? So all of this sits in these flywheels. And again, you can create it yourself or part of the AI brain. All of this is done for you when you do the onboarding. But right now, for most people, the reality is very different. So if you've been watching the last— this is like video number 3 of this series. You kind of see where we are going with all of that. So building AI so you can— building your understanding of AI so that you can go natively directly to the models. But the reality is most people right now, they sit in this place where they look at Instagram, they look at LinkedIn or whatever, and all they hear are wankers on the internet telling them like the 100 tools you need, the 120 tools you need to do all sorts of shit you don't need in AI. So I don't want more tools. I want to do deals. I want to be happy. I want to get progress in my life. And there is just this overwhelm of tools out there. And I know people who are stuck in this loop. And to be fair and honest with you, I went through this rabbit hole as well. I was like, oh, you test another tool, you test another tool. It's nice and great, but you don't need any of that. You don't need 120 AI tools like these guys would have you believe because None of them, when you look behind the curtain, none of them, there's nothing to it that you can't do yourself quite simply. So all of these are wrappers around ChatGPT or other models. So they basically, there's an interface that uses a model that you too could use. You know, there's no difference in the model they use. It's not smarter. Okay, what they have though is a context, you know, so they specialize the use of this model pointing at their context to do a specific job, you know. And then that— this is, this is really the key here. And you get to, you get to access all of this behind a paid subscription, behind a front door, you know, over there, you know. So they honestly— like, there is no tools out there— if you— there's very, very few AI tools out there that don't follow this model. So once you understand this, you understand that we are all here. And it's worse than that. All of these tools, they are great until they're not so great. They're great until another tool comes along. And it's just the forever chase. So all this time we move from migrating from one to another, getting used to one interface and another, if we do at all, because most of us also, we are just just stuck in just the FOMO and just the general overwhelm of it. For those of us who are hoping from one tool to— it's just, you can't, you know, we can't keep up, you know. And the reason— and yeah, they come and go, they have a shelf life for like 6 months. So you can't really invest yourself in a lot of them tools because they get disrupted by the next one and the next one. And at the end, none of it matters, you know. And we get stuck here paying the ignorance ignorance tax, you know. So we pay a monthly fee for some tools you can build just by talking to your AI, uh, just natively, you know. So that's really the idea, okay. And just have a look, if you ever look for tools out there, uh, in the age of AI, you know, like, uh, what are the ones worth you keeping, you know. So I'd say, so think of, um, Think of tools that give you stuff you couldn't have otherwise. You know, so the first aspect is obviously the data. You know, so without you know no point intended, but like the you know property filter and the property filter MCP API, you know you cannot get access to the motivated seller data. You can get access to all of that without it. You know, so if you you cannot rebuild this, you know you couldn't rebuild this. Just looking at creating some tools that go on Rightmove, like it's just unrealistic. It cost me 4 or 5 million to build over the years for you to get it for a few hundred quid a year, 300 quid a month basically. Like it's really quite unrealistic, you know. Think of other things like that where you've got like proprietary data that you wouldn't get otherwise and are meaningful and valuable for you to achieve your outcome. Keep a tool that has got a proven blueprint, a proven methodology to it that you cannot replicate otherwise. So yeah, think of if it's just data that's available on the internet, if it's just a wrapper around ChatGPT, which you could have as well, if it's just good because it's got a little bit of context, yeah, it's not really worth it. So there's lots of things around posting social media posts or whatever. Now your Claude can do this straight away, you know, so there's no added value in— all of these tools are redundant because now your Claude Desktop does this, you know, you can schedule all of your content, etc., through it. The plumbing is an interesting one. So for example, so Granola is an example. So you can't really— it's unrealistic to go and create or recreate a tool that sits on your phone as well as your PC, that syncs everywhere, that's got an API connection into your AI, that you just tap a button and it records, transcribes, gives a summary of everything. And it's just super useful to— super easy to use. These tools are quite a lot of utility. I call it the plumbing. So these are things that you'd be mad to rebuild. I wouldn't try. And lastly, it's like, is like also the people. So what is it that you subscribe to that's giving you like an insane amount of support, you know, or you get unstuck? There's beyond just the tool, there's a lot more, you know. So again, I feel like I don't want to make it feel like I'm plugging PropsFilter, the PropsFilter community, the 8 live calls, the live chat, you know, 6 AM to 8 PM, 7 days a week, all your questions answered about property, et cetera, and the software. Or whatever. Like, this is like, think of keep things that keeps you moving, you know, like community is a great one. So you'll find that a lot of tools, they are doing like what we've done for the last few years. They are adding, you know, a big community and support aspect to this so that you get really quite a big benefit from that. Okay, so, you know, and if you look at like technologically what I'm talking about, and I've shared this with you in the last couple of calls, right, these are just text files, you know. So in terms of like technology robustness, yeah, it will, uh, it will outlast your, your computer will burn, you'll still be able to retrieve, uh, the, the text file out of your machine, you know. Um, you know using a software, onboarding yourself on a software, the software is going to be redundant by something else. These text files are yours forever. And as you build on it, and again, go back to the previous video, I really show you behind the scenes how all of this works. You can see how this will outlast anything and everything. So especially now we think OpenAI and the likes of OpenAI, Anthropic, they're attracting most of the investment in the world. right now. Like those 2 companies that have attracted 50% of the investment in the technology company last year, which is crazy. So in the last sort of 12 months, which is crazy. But for me, they are just the AOL of the internet. So today's AI giants, they are the AOL of the internet. The share of at-risk investment capital going into the at-risk capital. So what was it like? 67% of the at-risk capital in the last 12 months has been allocated to AI, you know, which is crazy, right? So there's no investment in health or other stuff, right? Or other types of startup. And the majority will die. It's just the nature of it, you know. So now we think OpenAI and Anthropic and others, you know, they are Nice, great, and they're going to be here forever. If you look back at who were the big companies in the late '90s, the beginning of the 2000s, or even before, like, you know, Facebook has only been around since like 2005, you know, or something like that, you know. So Meta didn't exist 20 years ago, and there were loads of bigger, larger companies before that, you know, that don't exist anymore. So, you know, The battle is not one of, you know, like, it's not the one of which tool you have, which model to use and be partisan about it. You know, all you have at the end is your context if you choose to build it. And I'm a big advocate and partisan to that. You know, as a company, I could have taken a bet to say to myself, you know, us to make Prophecy Flow a wrapper. a wrapper around ChatGPT and get you locked in the user interface, get you locked in basically holding all of your data like everyone else does. But no, I'm taking the bet that this is the direction the world is going. The world is going in a place where people are talking to their AI interface. Their AI interface is connected to all the other tools they use, connected to other people's brains and frameworks, like your brain connected to someone else's brain, and all of this is going to work like that. So I'd rather be part of that transformation. I help educate you on that instead of lying to myself and trying to keep you locked in and make you pay more and keep you stuck in Prompt Filter, or, you know, for example, you know. But so that's, that's the idea. So because the tools will always change, you know, like if you see the pace at which those companies come and go, uh, it's— you don't have time. It's impossible to keep up. So It's a lost battle anyway. So, so do not be overwhelmed. Do not have FOMO about which tool to use, which tool not to use. It doesn't matter so long as you proactively build your context. And again, I redefine this context. It's 2 types of data. It's the niche. It's owning the niche data about what works right now in property. So every time you interrogate the Property Filter MCP API, your brain learns and just keeps it there. You know, I don't mind. Happy days. And, and secondly, it's your data, the stuff that is proprietary to you, the way you work, your goals, your constraints. And only with these things can AI give you an outcome that is above average and actually give you an outcome that gives you results, your utility on the other side. So once more, it's not about your implementation of AI, the way you're going to outwork teams of 10 people on an hour a day. By still being full-time in employment. It's not about which AI tools you buy, it's about the context you own, the context you create, you maintain, and you compound. That's really what it is. Okay, so in the next video, we're going to talk about what it actually can— what it actually does for you. So make sure you watch this next one. In the meantime, I'm going to put a link in the description here to an invite to our AI bootcamp on the 2nd and 3rd of September. And there I will go through the whole structure in great detail, the whole library. And when you register for it, you straight away get access to a 60-second test that basically tests your current AI against the outcome you want in property. And you know straight away, you know, how good you are doing and all that. You know, what we find so far is people get between 2 and 3 out of 25. And it's a big wake-up call. So people are kind of getting on board with this philosophy I have and I'm sharing with you to actually want AI to do stuff for us. So click the link here, join. The link will invite you to register to this bootcamp, 2nd or 3rd of September. The 2 sessions are the same. It's just people are free lunchtime, others in the evening. And then at registration, you get this test. You can put into your AI and see where you are. Okay, you get it all for free when you sign up to the AI Bootcamp. And once more, really the big idea, you know, if you want to be the smartest people, the smartest person at the dinner table, at the Christmas, you know, table, you know, AI is here to stay. And all the stuff I'm sharing with you is like, will help you be a top 0.1% implementer, you know. So It's really not about the AI tools you buy. Get everyone else to get FOMO and overwhelmed and stall where they are, complain and be fearful about it. It's about the context you own. And you own, as I said, you create, you maintain, and you compound. Okay, so that's what I wanted to share with you guys. Let me go and have a look at the Q&A. There we go. Oh, camera. In my camera here. Let me have a look at the questions. Okay, so if you're fully digital with no paper, how many levels of backup do you use? So I've got all my brain sits on my computer, and once a week, and once a week it's sort of saved on my drive. That's it. Any other questions, guys? Put it in the chat here. I'm going to— so that we we make the most of the time we have together. Questions, reactions, you know, just make it as interactive as possible. So let me know, Suzanne, if that makes sense. You know, I'm a very paper guy. You know, I used to journal lots. You know, I've got like stacks and stacks of notebooks and notes, but it was going nowhere. You know, and the reality is I take these notes because I'm listening in to a conversation, I'm in a meeting, And now I can really focus on the meeting and the creation, the idea, and making things progressing and be fully here. And the AI basically will just populate my to-do list. And then it helps me plan my week, plan my days, and really feel what's urgent. Yeah, really feel what's urgent and gets me focused on what's the most important. And I, you know, if this is the one thing it did, regardless of all the other things about the context, but just manage my days, I can produce probably 4 or 5 times more than I did only a few months ago, just by having this assistant, if you call it like that, my property AI brain that knows what's important to do and also challenges when I overcommit. You know, it's very easy for us to say yes to too many things. And then it's like, well, the deadlines are here and we can't make it all work, you know. So it's been really, really good to do— basically commit to fewer things, but like, these are the top 3 things I need to do today, the top 5 things I need to do today. And then, and then the compounding effects of actually doing what's the most important every day, uh, you know, there's loads of concepts around that, you know, like The Slight Edge, uh, there's loads of books around that, you know, like The One Thing. The one thing, the slight edge, really all of these, you know, we know we need to do these things, but how many of us actually do it? Thanks, Guillaume. Yes, that did answer the question. Thank you. But I was just wondering, does it mean that your context is stored both in the property filter system and on your computer? No, nothing goes to property filter. Yeah, so This is the bet, as I was explaining. All of this, if you looked at the previous video, so Claude Desktop is an AI interface that uses the Claude models, and it points at your context, it points at your folder on your machine. So today I've revealed, you know, the structure of the, the structure of the folder, you know, the utility base, and you own all of this. And tomorrow you can ditch Claude And you use another AI interface with another model, you know, a Chinese model, ChatGPT, or Mistral, or some other, whatever, right? DeepSeek, you know? And you point it at the same, your context, and it will go and perform the same way. Tuesday, so Tuesday was the last session of, so we did like a first cohort. We had the first Props Filter member who got their brain in, you know, their property AI brain installed. And just a few days ago, we had the last implementation session of the cohort, you know? And like I always do, I'm last minute with lots of stuff. So I have AI to do my slides. And because it's the last session, I wanted to ask my brain, my AI brain, I just did stupid things. So basically I went through all the transcripts of all the calls I've run for them against the curriculum I had at the start. And I wanted to say, Can you find if there's any gap, something I've not talked about to make sure I've been exhaustive and I've covered all the stuff I wanted to cover? And I've done something stupid where I've copied like 6 hours worth of transcripts or something like that, even more than that, 8 hours worth of transcripts. And I just made my AI interface, you know, I should have put it into a document rather than put it in the terminal, right? Basically. And I made the whole thing break. And it broke. I couldn't start my Claude again. And I was like minutes until the last session I had to deliver for the community. And I had no slide. I wasn't sure what were all the stuff I had to talk about, et cetera. So I went and basically I asked my AI to fix it. So through the phone, ask me what to do. So I went and disconnected basically the, whatever, the Claude folder on my machine. And so it restarted, it managed to restart Claude, but it was like a virgin, you know, totally, you know, it was like a brand new, I didn't lose anything, right? I didn't lose anything, but it was like a brand new onboarding into Claude Desktop, you know. And the same way, if I used, if I'd use another, if I would have opened a ChatGPT or something like that. But because I own my context, I was managed to point this to that folder. And it did my slides within minutes, just like nothing happened. So I didn't have to prompt, tell him who I was, what I was trying to do, what was the program about, etc. So from a brand new, if you like, a brand new, you just start Claude for the first time, I pointed it at my context, which has many, many months of all my work in it, and it just done it like nothing happened. So I'm sharing this example to show you, to show you that when tomorrow Claude goes out of business, you know, and you want to use something else, you know, all the— all that's left is your context, and you own all of this on your machine. And me, as Props Filter, when we basically support you and provide you this brain template, it's just a folder on your machine. And you have a connector inside the Props Filter MCP that gives you all the Q&As. All the coaching you can have from the community basically. And then access to your pipeline, access to your search and property filter, all of that. So it's kind of like one way coming to you. But in all your contexts, you own, I mean, yeah, they've got no interest in your stuff, right? So that's really the idea. So think of, this is really like, as I told you, like a top 0.1% AI implementation, you know, because you really and truly understand how to go direct to the models. rather than being bussed around between all the different tools there are. All them tools are redundant the minute you understand that you can own your context and therefore make the AI models do what you want to do. Okay. Thank you very much. Yeah, if that makes sense. Yeah, cool. Do you mean this person will be stored? Okay, so we talked about that. My writing, I am drawn in paper notes. What did you use to transfer them into your brain, please? So you can, I would say you talk to it, you know, so you create a session. So you'd create a session, you start a session, you know, on, you know, like you start a chat and you say, okay, so today I want to talk about this course I was on, you know, and you can just walk through your notes, you know, like this was the big idea, this is the benefits, etc. And I would talk to it, you know, if it's all like that. And it's a good— if it's valuable to you anyway, to, you know, if it was valuable to you, it's always good to go over your notes anyway. So as you do, you do it out loud, and then this knowledge then compounds, you know, if that's relevant. And you'll find that, you know, if you're like me, like a lot of the notes you've taken are redundant with newer notes anyway. So you start from the end. And you find that if you redo a to-do list every other day, like I used to, well, after a while, none of these things really— none of these things. But I would do like that because the context of the exercise is about learning knowledge. So I think if you had a robot that comes around your house and takes your folders and then scans everything, Uh, you know, yeah, you, you'll get the stuff digitalized, but you also do it to learn, you know. So I would, uh, I would do, I would do that personally, okay? Especially if it's on return and stuff, okay? Uh, David, I have a VA doing, I have a VA doing 100 hours a month across the whole business— the sourcing, assets management, bookkeeping, monthly financial reporting, social media management. Do you see the AI brand replacing the VA or augmenting the work of a VA? Absolutely. So I'm telling you, like, the impact this has had for us. So me and Florian, if you know, like, the other co-founder, we used to work 7 days a week just to keep the plate spinning, you know, like 10 months, 10, a year ago, you know, and literally could not take any initiative. And we had a team of 42. Now implementing this, we're now a team of 28. I've— when was that? 4 months ago, I gave a budget to my board saying that we are freezing new hires because we are going to do things with AI first. And the only reason we hire people is if there'll be significant opportunity or to do something else. And yeah, so that's the reality of it. And then we went from working 7 days a week on the business to 3 days a week on the business. So I get to Think through things like that to share with you and taking initiatives, support you better and things like that. A lot of the team has been retrained. Everyone's using AI. Everyone's doing a lot more things. That's the reality. So in your use cases, so deal sourcing. So you've got PropFilter with when you connect the MCP to your AI brain, in PropFilter you will have set up like lead generators, your saved searches. So you'd ask, you can run a command every morning to say, you know, deal flow, you know, or slash deal flow or whatever you want to call it, slash deals inbox. And it will tell you like, these are the 5 viewings worth your time. Okay, cool. So next step, you might go and have a look. If it's a utility, yes or no. And then you go and organize the viewings. You know, like people think, you know, like people tell me like a property is hard and stuff, but they are just like scrolling through pages and pages of properties, you know. And the reality is you need to look at the dashboard or you need to look at the information for 2 minutes a day, 5 minutes a day to know and take a decision. And decision is pick up the phone, we go and view that property. So people call themselves investors, but really they're just cocking around, looking at their screen. They're shit scared of the phone. They don't want to do it. They're lying to themselves, lying to their spouse, lying to their kids. when they say they want to be property investors and they don't progress. They're just on their screen doing nothing, right? So all of this makes you progress all of that. And then you find that your next job there, once you've identified this, is, okay, I've been on the viewing. You just talk to your granola, for example, that then speaks back, then brings back the context into your brain. Literally about take viewing notes about, you talk to your phone, So okay, this is this. This is the situation of the vendor. The agent this. It's on for three seventy. They told me they had an offer for three thirty. Like you know before you're home, all of this context is is ready. And if you want, you can create a routine where your your AI creates like a comparable comparable report that you can and and and draft an offer in your emails directly. And then you're ready to send based on all the conversations you've had, you know. And so it's just all of a sudden you can outwork people who have traditionally a team of 10 people on an hour a day. That's really what it is. Asset management. So same thing. So you'll have in your flywheel, so your portfolio, you'll know for each property, the renewal dates, the interest rates, you know, like the, you know, all the admin, like people don't realize, you know, like the all the admin, the compliance of running a portfolio. It's so fucking noisy, like the refinances, the insurance renewals, the optimization of insurance renewals, just the back and forth communication. When you go and renew your insurance and you have to look for, oh, where was— is there a flat roof? Or you say it once and it's captured forever, you know. So all of this admin goes away, you know. So you don't have to do all of that. Bookkeeping is the same. So So for— we've really gone low-tech, you know, with bookkeeping. We used to have like complicated softwares you scan through and whatever. Now it's all through the emails because the, the CLO has got access to the emails, goes and categorizes it in the right place. And obviously you still go and do the, you know, you still go and do, uh, make sure it's, it's, uh, it's— you just verify the ledger. But the cool thing is once it's learned to reconcile things a certain way. In the main, you know, once you're in business for 2, 3 years, there are no new expenses, you know, like, you know, so it's kind of always the same, 90% of it, you know. So that just all my stuff across all my companies, like it's crazy, but all my receipts are reconciled like within the week, you know. I get a little, I get an email, a message in Slack with like, these are the missing receipts from last week. I'm the one who creates most of the mess with receipts anyway. And then it's just there, you know, it finds it in the email when they're in the emails, et cetera. So then your VA, David, you know, can go and do a lot more interesting things, you know. So for us, the way we've implemented this AI brain is it was initially with the the founders, then the senior team, and then the more competent people in the team. So I think we probably have, I don't know exactly, but probably 8 of us have a brain and like a personal AI brain. And then whatever assets we create as we, because anytime you do something. So I'll tell you a thing about SOPs, like about systemizing your business. There's a There's a— I'm gonna have the E-Myth, you know, over here. So, um, you know, you all know you need to have systems and process and stuff like that. The truth is you can't sit down and say, okay, okay, this is how we're gonna go, this is how we're gonna do business. You know, you don't really do that. The best way I've learned over the years to write SOPs is to write them as you do the task. And even knowing this, the reality is you're busy doing the task, right? So you don't record your screen, you don't record yourself, etc. Now, so I don't write, I only write through Claude Code. I don't write into my emails, I don't write into anywhere else. If I need to send something to a team member, I literally start a session on my phone with Claude and with the remote control. And I would say, can you send this to that team member? Can you write this to that team member? And it will go and write it on Slack. We'll go, if I need to send emails to people, if I need to set up an appointment, can you book me a train on this? You know, it's just like that. So that's really the reality of it because I do everything through it. So it records all that I'm doing. So I'm like, okay, great. So can we make a skill of that? You know, can we make a skill? And I want to run this as a routine every day at 6 PM. Then that's it. You know, so every time you do something, it's an asset for the next time. So, so yeah. Uh, yeah, so, and then it means again, David, that doesn't mean you have to— we're going to sack everyone, but, um, uh, your VA can go and do more interesting things, you know, if she can, or she can, or he or she can be trained. That's, uh, really, uh, really, really, uh, really, uh, really cool. Uh, Serena, AI Brain helped me move bulk email from Gmail into R Drive using, uh, Google App Takeout. Yeah, absolutely. The thing is, now the— there's no limit to what you can do because is the only technology who, if you don't know how to do something, you just ask it and then you just follow the steps. I don't understand most of the stuff it makes me do, but it works, you know. So that's, uh, that's the, uh, that's the idea. So same thing, my emails, for example, I'm close to inbox zero. My personal email, I used to have like 23,000 emails in there, uh, like unread email, you know, and, uh, or something like 29,000 or something. And now I've got like, uh, the top 5 I've got like 3, 5 stuff that are current right now to do with like purchase and sales of properties. And I'm like chasing people. And then my professional one, you know, as well. So I've got my different emails connected to it. It's all in there. So the AI brain suggested takeout where I told him what I needed. Then the brain sorted the folders out with the hard drive. Yeah, so the question is for about the OCR. Yeah, so then soon I'm going to make myself redundant, you know, because you're just going to just go and ask it, you know. So, and a lot of it is mindset. So because we don't know all that it's possible and the mindset of asking the AI instead of asking someone, it took me months, you know, to get used to it. I used to be that little pest, you know, to ask Flo and my other co-founders, you know, how do I do this? How do I do that? And then I, mate, you know, like you're just as good as me at asking your AI, you know, so ask your AI. I said, okay. And then, you know, you can, you go and you truly transform your identity into a, you know, becoming a problem solver, you know, of all things. Okay, so hopefully that was useful. Hopefully that was useful. I like lifted the curtain on the organization of this folder again. So the big idea is if you remember, your context is really your, The data that is about your niche, the stuff that are working right now, if you can tap into it, and everything that's about you. And without these 2 things, it's really hard to do anything above average. And you are stuck working for AI because it's going to be really quick at giving you average stuff that you have to copy and paste and edit and do whatnot with it. So yeah, everyone's got about the same as me, about 23,000 emails. So yeah, you can go sort it out. So my AI has gone and put label on all the newsletters. Labels on the stuff that I kind of want to search. You know, I've got FOMO if I don't have it. So put a label newsletter, but skip the inbox. I don't want to really see it every day. You can go and manage all of your inbox like that. And then for like every day. So instead of looking at my emails, like I don't have my emails, you know, I don't look at my emails like before. I would look at it every, I don't know, every 20 minutes or something. I don't care, like twice a day I run a command that tells me like, here are all the emails in the inbox. I suggest we archive this. I suggest we need to respond to those 3 ones. This is a draft response. And I talk back to it and then we deal with everything. That's really the idea for personal as well as the business ones. And if you have like 15 different email addresses like me, you can obviously do all of that. Okay, I hope that was useful and I shall see you I shall see you next week. Thank you. Speak soon. And in the meantime, so remember, like, book yourself, like, look at the links in the description and book yourself for the AI bootcamp if you're not done already, because we're going to go through all of these subjects in a lot of detail, 2nd and 3rd of September. And if you want to assess how well your AI is doing, as soon as you register, you get the test back that tests your ChatGPT. Whatever you use, and you'll see, uh, you'll see how good, uh, uh, you, you are good or far away you are, and, uh, and it will help you, uh, know what to do next. All right, speak soon. Bye-bye.