The 3 Levels of AI Mastery

Some people it makes them cry because it's just how does this know me so well? They're like, oh, this knows me better than my spouse. And then I asked it, what am I most afraid of? And then the five personas all answer and one of them popped off with, they said, what you're most afraid of is quiet irrelevance. You have the library of Alexandria within a web interface that is freely accessible, maybe a $20 a month. And then people are telling me that they're not using it regularly. Are you insane? What it means is just chat chippy tea with somebody pressing the buttons. Let's just think of it as another AI using the chat interface for you.

You can't even write an email reliably 10 out of 10 times. So like, what are we talking about? And yet when you talk to the people that were in the trenches at the beginning building them, the way they talk about them is they are not programmed, they're grown. It's not about replacing. It's about enhancing. The next day. Igor, let's, this is about you today. And everybody wants to hear what your thoughts are on the cutting edge of AI and where things may be headed. And I wanted to just start with, so like, what's the most exciting thing happening in AI right now that not everybody is really aware of yet?

The most exciting thing. I mean, where do I begin? It's just moving at light speed. So I think first, I would just like to put it into perspective because there was this one thought that I had recently. And I think people need to realize we're moving. We as a society and all these tech companies, we're moving towards this idea of artificial general intelligence or just an AI that is smarter than any human and can work autonomously. We're moving towards that goal. There's many definitions. But like, what we get to do as consumers, as individuals, is if you picture it as sort of like the landscape being like an ocean, and then artificial general intelligence is kind of the destination somewhere beyond the horizon, then all these tech companies, they're these big tankers, these big, big ships just going through the ocean kind of going into different directions.

But what they do as they go in that direction, is they create a massive wake, create a big wave. And then I think of myself as just being blessed enough to be alive today and have access to the internet, I can surf that wake of the big tankers and have not just a good time, but also win. Like, win in what I'm doing. Win in my personal life, win in my business life, and everything in between. I can spin up projects at the speed that is unbelievable. I had an idea. Let me tell you, I hosted a session today with some students that are really at a high level and building things, creating things. I had an idea 30 minutes before the session.

I was literally in the Zoom room in the waiting room, and I spun up a website with, it was like a directory for all the submit, we did like a challenge. I did a submission challenge homepage with 20 cards with custom visuals with multiple layers of filtration. It took me 10 minutes. It took me 10 minutes. And I showed up in the session, I was like, hey, I just made this. And then I was like, I caught myself like, wait, I just made this in the waiting room of Zoom. Like, this would have took a month with developers, a lot of money back and forth headaches. It's just done. So, so I'm just grateful to be doing all of this.

And I think the most exciting thing for me is the ability to experiment with this new stuff and reteach it. I think ultimately what you asked for there is kind of like what's one thing that people should be excited about themselves. And I think it's this change. It's the shift from a chat interface or a chatbot that chat chippy T is and has been over the past few years into even a lightweight agent that just does things on a schedule for you. I think that shift, if people become conscious of that, well, it's incredible because all the stuff, all the chats that you might have laying around in your chat chippy account from the past few years, anything useful.

And we could talk about that. You can dig it up and you can set it on a schedule and it just happens once a day. And that's, that's like something that consumers weren't able to do a few months ago. And it's a reality now and it's available at like $20 an hour. Okay, let's come back to that one. I wanted to just jump on the thing you mentioned there about spinning things up nearly instantaneously. And I had an episode recently with the CEO of Zapier Wade Foster. And he when I asked him a similar question, he said, the cost of coding is going down to this tiny, tiny level where it basically for someone who's curious about this stuff and wants to get involved with it, it really only costs your time to build something that in the past you could not have done before.

And he said, I don't think the general populace has wrapped their heads around this at all. And so I do feel like that's a very exciting frontier. I mean, I know I currently have a project going that I wanted to do maybe five years ago. And I was quoted by a development team that they were they said, this is probably going to cost somewhere around maybe $200,000, but it could be more than that. Because it's going to take like at least a full year to build out. And now I'd say that project is 90% of the way there. And I've done it completely on my own, which is just mind blowing. It's yeah, people don't realize how good it has become.

And yeah, developers are scared for reason. I mean, it's just, but the thing is like the good ones, they've just become like super human because they're just turning out projects after projects. It is an exciting time. But there is a skill to it, right? Like, I'm sure as you worked on that project, Nick, like you learned a lot in the process of working on that or other projects before that, like it's not kind of just you don't you don't get good within minutes, you get good within like 20 to 30 hours, I would say in total. So and that which is a ridiculous learning curve compared to spending five years studying computer science and then going into a startup and working there for a few years to get to a level where now people are in two hours, right?

But but there is a bit of a learning curve. Yeah, I don't mean to overstate how easy it is because I think you still need a you do need, it helps to have a basic understanding of say software architecture and not just come into it having done none of this. And then it also requires really a good understanding of logic, even if you don't have the coding background, you have to be able to think through things very sequentially and logically to get the models to do what you want them to produce. So the point well taken on I don't want to overstate how easy it is. But if you're willing to like, yeah, that's something complex what you're saying just for context for anybody listening, right?

If you're doing something simple and want to spin up a one-pager directory like I did like my grandma can do that. Yes, yes. And even that is very exciting. Now what was the chat GPT thing that you were describing there on the you know, on the $20 a month plan that we can go back and use our old conversations and. Yeah, so no, it's just this general idea that, you know, if you've been using chat GPT over the past few years, you probably have some useful conversations in your history and they don't get deleted, right? They're just sit there. And one thing that I've myself and for others found great success in is just going through those chats and asking myself the simple question is like, which one of these would make my life better if it ran again in the next week?

Right, just very simple, you know, you get to define better. But if you, you'll find that some conversations are going to be one of things that you explore in a moment and maybe, you know, you're never going to do again, you'll find that there's another category of things that were helpful within the context of a project that might be concluded at this point or were also timed to the timing. But there's a category and it's probably, you know, it might be a single digit percentage of the chats you have. But there's some chats in there that would make sense to be re-ran another hundred times in your future automatically.

So I don't know, like a typical example that I see a lot is content creation. Like people, you know, and people, this is just how we are. We often do a thing one time. And then we're not, you know, we didn't make a habit out of it. We didn't build a system. We didn't put it on a schedule. If we did that same thing a thousand times, we would have a business and be crazy profitable. But we just do it one time and move on to the next thing because that's the, you know, next bright thing. So just going back into your chat history is a very practical thing that anybody with a little bit of chat, chat GPT usage over the past can and now with these agent products that are very accessible.

Talk about that. You can just put it on a schedule. You can be like, that was great. Now do that every day, three times a day. And it's not going to cost me any extra dollars. Like, that's just, that's what I meant. It's really powerful. Yeah. Well, let's, let's run with that. Where are these agents at? I mean, I think most viewers, listeners, they understand what an AI agent is, but not necessarily always how to apply it, how to make the most out of these tools that exists. Like we hear this agent, like AI term thrown around a lot, but you have a real expertise in actually taking these tools and then making them work for you to achieve specific outcomes.

And so I'd like to dive into that a little bit. No. So the problem with the word agent is you're going to ask 100 people, you're going to get 100 definitions. They're going to be similar, in some sense, right? It's going to be like an AI that does things or an AI that connects to data or applications or an AI that can work on its own, right? They sort of overlap, but like really, what does that mean? What it means is just chat GPT with somebody pressing the buttons. That's at the base level, right? It can go further than that, but at a base level, it's that you're not the one copy pasting the prompt and adding a little bit of text.

It's this agent doing it automatically for you. So you can think of it as one chat GPT sitting behind a computer with a keyboard with a chat GPT account and then doing things, right? That's where it sort of begins. And it's also like really easy to visualize that because it's just a natural evolution of you using the chat interface. All of a sudden, just think of it as another AI using the chat interface for you. I like that. So that's kind of the simplest definition that I can come up with. Now, where are they? There's so many places. And that's a confusing thing. Every marketing company, every company's marketing department kind of latched onto this term AI agent.

And now you'll get, I don't know, you'll get a smart fridge that is agentic. Like what does that like? It's just or like all these products, honestly, every image creator, video creator, every website builder has an agentic builder in it now. What does that mean in the context of a website builder? It probably means that, you know, before you could use something like chat GPT to create the website. But now you just press a button and it prompts itself and now it creates the website with chat GPT for you. That's really what it means. And maybe it doesn't run one prompt, it runs free, right? And they're like, it's agentic.

We can market this and sell it in a big way and raise at a higher valuation next month because now we have agentic AI built into the core of our product. It's like, man, you really, you have free prompts that run themselves inside of chat GPT. Right. And then it gets over hyped. They make these hype campaigns. Like one thing to realize, and this is a bit tangential, but I do want to say it is that all these companies, they're incentivized to grow and to raise more money. And what does that is, you know, getting more eyeballs, getting more consumers. So they over hype everything. They like, they take the hype lever and they go all the way, they're like, Oh my God, if we get it now, we can raise out 200 million instead of 150.

And then we as the consumers sit there and we're like, what? What they do that to really? It's crazy. And this goes even for the big companies, right? They're like, here's a new model. Here's a new product. Oh my God, you need to use it now. This is going to replay the future of work. And then you try, then it like works two out of three times. Then one time it fails catastrophically. And then you check the other two outputs and it made mistakes to and you're like, what is this? They hype it. So in reality, agents, chat GPT using itself and where you can find them, I think if I were to give a singular answer right now, we're like approaching, we're like almost in June, 2026.

And I think right now, Claude Coerk is the one application that is sort of the sweet spot to use agents for consumers. It's literally just if you're familiar with Claude, if you're not familiar with Claude, it's the same as chat GPT, but it sounds a bit different. And then if you're familiar with Claude, which you are now, then it's basically Claude with access to a folder on your computer, you can give it a folder. And it can create files within that folder, it can delete files within that folder, it can create new folders within the folder, right? And then also you can put it on a schedule. So if you're like, hey, you know, search the internet for these three keywords, you know, AI generative AI news, you know, today, and then take the top five results and turn each one of the results into a little tweet and save them into this folder.

It just does that. And you can schedule it to run every day at 8am and you wake up. And there's five new tweets for you based on today's new stories. That's an agent. It can go deeper than that. Sure. But at a basic basis, that's it. Chatchifty prompting itself with access to a folder on your computer. Yeah, and there are different platforms where you can do all of that. I am in full agreement with you though, that where Claude is a step ahead of everybody else with the co-work functionality makes it much easier for people to build these things out and get immediate use out of it right away. And Claude Code is, for my money, well ahead of the other model's ability to code right now.

Although I do hear Google just announce, I mean, they have they're moving in that direction too and have a number of consumer facing agents that are about to be flooded into the market. Will they be over hyped or will they really make a difference? I guess we'll have to wait and see. Yeah, we shall see. I mean, I guess there's like really four competitors for that phone right now. If we're honest, there's OpenAI, which they have, they call it Codex, that is their agentic product. Then we have Enthropic with Claude and Claude Code work and then Claude Code, as you also mentioned, which is a bit more manual.

I think co-work is the sweet spot, but I would recommend to eight out of 10 people. And then we have Google's, they have anti-gravity, but then they have also a jet. I won't even start because their product lineup is a bit of a mess. We could confuse everybody very successfully within 20 seconds here just by naming it all. They have a few things going on. And then Grog also came out with sort of a command line tool that is usable to, I think, those that are main four when people have these conversations. For most people, potato potato, just pick one and run with it. And you're gonna have a good time.

It's actually really fun. And people forget if something doesn't work or if you get stuck or if you're just staring at the blank page, it's still a chatbot. You can still kind of be like, what do I do? And then it's like, oh, yeah, we can do this. Right. Right. Absolutely. Now, you were tell me about your school, your program, your training courses, because you mentioned earlier, you just finished up with the mastery group and had an amazing experience. Let's dive into all of that. Oh, yeah. So we do a lot, but my base philosophy is this. I think AI education is futile if it happens in isolation.

I think you need kind of, you need a map and you need motivation to do it, because we can all agree that learning AI is a good thing. Right. That's an easy thing to agree on. We can all agree that it's going to be only more relevant in the future and already in today's landscape and certain knowledge jobs. What are you doing if you're not using it? If we hire new people, I'm going to be asking like, what are your prompts? What do you use? Do you have project? Do you have do you manage context files? Right? Like these things. So we can easily agree on that. But then the reality is that most people like, they're like, yes, yes, yes, they're like nodding along to me saying that.

And then the reality is that two months pass and they haven't tried a single new thing and not learned a single new skill, and they just keep hearing about it. It's really easy to consume content like this of, you know, to the handsome young guys just discussing AI on the internet. You know, it's really easy to do that. But it's not as easy to do more. Let's focus on that answer. Let's give it the conversation to us. Continue, continue. So it's really, really easy to say it, but it's hard to do it. And what we found is that you need to be in an environment and an energy and in a community of people.

So whether that's live or it's an online community, but this stuff is just really hard to do on your own. The knowledge is out there. You can find that you can piece the stuff together and watch 50 YouTube videos, but they're going to be disjointed. They're going to be from different voices. And you're not going to have one, you're not going to have the motivation and two, you're not going to have that map that you can progress along. So what we did is, is high level what I did, I laid it out after seeing hundreds of thousands of students go through the stuff. It's always the same thing. Come in and it's sort of like this pyramid.

It's free levels. You can imagine it as a pyramid. The first level at the bottom is AI for answers. It's people using it intuitively. Asking questions, getting answers. So you start using it kind of like a fancy Google, but quickly, like you start like problems, if you think about it, are really also just questions you have. It's like, hey, can I write an email and then chat GPD responds? Yep. Here's the email and you're like, whoa, but really you're just kind of asking it questions, but you're already winning. That's, it's like, I'm not talking bad. Like that's an incredible thing to have at your disposal.

But people at that level usually feel like there must be more. Aren't people doing more with this? And they would be right. You can do way more because you can develop a depth of context, a depth of intimacy in terms of knowing you and your patterns and your goals and your values and the way you write. You can develop all of these things, but it takes a little bit of effort. And that's what I call level two. That is really using AI as a partner, as a thought partner, as a decision partner. Man, Nick, the way it kind of like sharpened my thinking over the past years is unbelievable. I wouldn't make a high level, an important decision anymore without consulting a context-rich AI partner that I set up over time.

It's just too good. You can explore other sides of arguments. You can in relationships, it's unbelievably helpful. You can recognize patterns as they unfold and course correct and just get better outcomes across everything. Another thing that is easy to agree on is just life is just a series of decisions and your outcomes are defined by your decisions. Why not have somebody who has read all of humanity's knowledge as a part of that decision and not in a way where it influences you. People are scared of that. But usually when you ask something, it's just like, well, you could consider this point of view or this and maybe you don't see this.

If you set it up correctly, it does that. So that's level two. That's having it as a thought partner. And before we jump to that, I'll just tell you that. Yeah, yeah. And just before we jump to that top level, one thought on that, like there is an art to that level two with strategy and whatnot, where if you have given the model, the proper context about you and your goals is a great base layer for it to work from. And then actively pushing the model to poke holes in whatever you're doing, right? Because they all love to tell us that we're geniuses and we're brilliant. And the new thing that we came up, oh, no one's ever done this before.

This is amazing. But if you get the model to flip it around and say, I need you to poke holes in this, where what am I missing? How would someone pick this apart? They will flip into that mode. And I find that's where it's most useful, where I can get it to be almost combative with me because that sharpens my thinking and opens me up to things I would have missed in the excitement of trying to create something new or, you know, move something in a new different direction. 100%. And that's the beauty of it. Like, as it knows, like, it is an art, as you kind of massaging these details about you, it adapts in a very intelligent way because, for example, from what you told me there, you might need that sort of advice.

But some other people might need the opposite. Some people maybe are way too harsh on themselves and they may need a little bit of gentleness in the process and then it will adapt in that way. But it needs the right type of context and it's not the type of context that will automatically be picked up by the memories feature, for example. It's a type of context where you actually have to sit down and ask yourself those hard questions and that why behind certain things and and like this is this is one thing that we guide people through our teaching, like exactly what are those questions and what are the sub questions that follow up on that and how deep exactly do you go on these certain things to make it a true partner so that it steers away from this default setting of just being a yes man on everything.

It's annoying and it's not helpful. That's level two. And then level three, I mean, pretty simple as we call it an architect and that's people who are building people who are building systems. So whether it's a gentic systems or they're just putting certain things on autopilot or they're building sites and vibe coding things or they're just building out really complex prompts that work in work with data that it pulls in. It's just people who are building systems. So they're not doing the work with it. Though if they are in tandem, right? Level two, you're still consulting and it's a partner but level three, you're kind of giving the reins over and it's running a system on autopilot and that is level three.

But the key thing to understand that the big mistake people make is they go from level one straight to level three and then they're surprised that the quality, the reliability is not there. It's kind of like hopping into business with somebody you just met. Or maybe even a better example, it's like marrying somebody you went on one date with. What are you doing? You should, what? You're going to commit to this person forever after meeting them for like, you need to develop a little bit of a depth of understanding at that second level, right? In a human partnership, for example, if you just met somebody, you're getting to know each other.

But a few years in, you know each other's patterns a little bit, right? And maybe even if you're lucky enough, you'll be able to predict when the other person is about to do that screwed up thing again where they always hurt themselves and you can kind of course correct a little bit and you can help each other out, help each other grow in all dimensions. And it's just that intimacy of knowing each other's patterns that makes it possible and it's no different with AI. If you know, if you, if your AI knows you and I'm not talking about your deepest, darkest secrets, I'm talking about the truth behind your goals, for example, but that's where it begins.

As I said, it is a bit of an art and just to go full circle. So this is the map that we take people through. And for example, we have the AI Advantage Club where we guide people through this progression when you come in and watch the onboarding, you become level one by default. You're here, you're doing level one. And then we have a course that elevates it to level two and guides you for this process of setting up these stock context documents. And yeah, I'm there explaining it step by step. And that's basically what we do. And as we go through that process, obviously, there's more there and we have more intense programs for people who want one on one attention or want to be part of a bootcamp of people, you know, being at this different levels or getting to level three.

But the base idea is like you got you got to unlock depth of context. No matter how you do it, no matter if you do it with me or somebody else or by yourself, you figure it out. Depth of context is the gateway to actually building systems, which then is the big leverage unlock in AI. Yeah, yeah, that's well said. I had a great conversation with Rick Nucci recently, who's the CEO of Guru. And what their firm does is connects everything within a large organization through a cohesive AI model. And it's interesting how similar what he described yesterday when we were when we were chatting to what you're describing now, he's talking about on an organizational level, but you're talking about it on a personal level.

But it's really the same principles where you've got to do the hard work of identifying your values, either as a person or as an organization and why those are your values. And that sounds like a simple thing. But when you start doing it and really go in deep with it, it's serious work. It might be simple, but it's still hard work to identify those things and really give it the context that everything else is then built on top of. And I think what you're talking about on a personal level, and he was talking about on a company wide level, is that if you don't build that layer well, that that solid foundation, when you try to move up to the advanced stuff, it's going to be garbage in garbage out.

Whereas if the foundation is there, now it's quality in and you will be amazed by the output. Yeah, it's and people, people have their minds blown. Let me tell you like one of my favorite things is if we when we run some of our bootcamps, as in the first session, we have them set up, we call it a personal DNA, which is basically a big form on on some of these topics in context on yourself. And even if people are sloppy with it, and they kind of brush over it and they don't go that deep, the thing is the AI, it has read all the literature, seen all the movies, read all the books, basically, right, seen all of YouTube.

So it knows human patterns. It knows if XYZ, then this is the consequence. So even if people are like super surface level and they say like, Oh yeah, like I'm an electrician and I have this goal and I want to do this and that it's like a certain type of person becomes an electrician or an accountant. And then if they're in this age bracket, there's certain patterns with it, it just has so many patterns on humans that even from the basic stuff, if you ask a question, I would challenge everybody in the audience, just go do this right now and chat Claude. I don't care where you add. If it has some context on you, just ask it.

Tell me one thing about myself that I would know. Yeah. And that's like some people, some people, it makes them cry because it's just it's just how does this know me so well? Some people, some people said this comment, I thought it was so funny. They're like, Oh, this knows me better than my spouse. Which is a scary thing. But that's at the surface level of context. So if you go a little deeper, it can you have the power of all of human knowledge. You have the library of Alexandria within a web interface that is freely accessible, maybe $20 a month. And then people are telling me that they're not using it regularly.

Are you insane? This episode of the next Stanley show is sponsored by granola. One of the worst things you can do during an important meeting is try to take notes during the conversation. You spend half the time typing half the time listening all the while not doing either one particularly very well. That's why I've been using granola. Granola connects with my calendar. So my upcoming meetings are already organized and ready to go. When the conversation starts, I can open the live meeting display and stay focused on the person I'm talking to granola captures the conversation and turns the transcription into structured notes that are very easy to use afterwards.

Instead of ending the meeting with a page of scattered fragments, I have the next steps decisions made and follow ups clearly laid out for me. I can also add my own notes and combine them with the context of the conversation for interviews, team meetings, and strategy calls. It lets me stay present without losing the details that I'm going to need later. Learn more about granola using the link below. Yeah. Well, along those lines, I mean, I did a created a AI board where made personas for a number of big time people. I mean, Seth Godin and the Volravicon. And then I also threw in like a Joe Rogan to have a just a completely different viewpoint.

And there were there were five of them in total and then took that GPT that had these personas in it and then connected it to all of the data from the YouTube channel, right? So it had transcripts of every episode, every conversation I've had all of the analytics. And it started with stuff like, okay, what's really working on the channel that I wouldn't necessarily have noticed on my own, right? Just digging into the analytics. And it had insights there, which were great. But where it got weird was when I started asking it personal questions, because they just started to now it was like a new level of intimacy when it had access to all these transcripts.

And then I asked it, what am I most afraid of? And it was it was almost like scary to ask the question. And then the five personas all answer. And one of them popped off with they said, what you're most afraid of is quiet or relevance. And it just like struck me because I thought, that's not what I would have said. If someone had asked me that question. And yet here is this silicon based web interface that is seeing an insight into me that I thought was really profound. And I just thought that that is an amazing, it's just that's an amazing development in the world of technology. I mean, amazing.

Those experience and you know, it's hard to go back once you kind of have that visceral reaction to something that it spits out. Everybody's like, oh my God, what else can it do? And there's a lot. So and it's that's yeah, powerful. Yeah. Yeah. Where do you think these models are going and what might be possible in the near future? We don't have to go like five years out just in the near term and what gets you really excited. I think what I'm excited about is reducing my phone usage in the future. And I think a lot of the stuff I do on my phone is trivial. And I'm just pressing buttons and looking into a screen and being disconnected from my environment.

I'm excited about my agent knowing that I have an a I don't know, a game like tennis game or a paddle game 30 minutes from now and that I should be leaving in 10 minutes and that it should call on Uber. But traffic is high now. So it's going to call one now and let me know. And I don't have to sit there and be on my phone and stare into this thing and be calling and canceling Ubers. I'm excited for that. Like the little things just reducing friction in the everyday life. I think the current interface that we have, which is like a super computer with a screen in our pockets and applications that are made for human thumbs.

I mean, seriously, this is the interface for the internet. Really this thing, like this fum, like that's crazy. And then like, I don't want to be staring into this thing. So I'm excited for in the near future. The next, I don't know, probably like year or two, we're going to get a lot more phones with small language models on there. And we're going to get a Siri that actually works. Right. And I actually, this is something that I would love to, I love to like see like just a Siri that takes a bulk of the applications that I use and does it better than me? Because what do I know about the traffic patterns in my city today?

That's all in Google Maps. I'm not going to be checking that. I'm just going to call my Uber and expect that a 10 minute ride is 10 minutes. But I don't know that there's a lot of traffic and it's 20 minutes today. Just do it. Just get it off my plate. So that's what I'm excited for. Yeah, that seems very doable in the near future. And even with my experience recently with Claude code, where I am trying to build something and it will say, Hey, have you thought about trying to get the data that you're looking for this way and going, No, and I didn't even ask you to look for another path. That was really insightful.

Yeah, let's try that. And then it works. I mean, that that little logic leap that you're talking about right there where it would start to infer the steps in between, right? I mean, it might know you need to go to your paddle game in 30 minutes, but it's those those little inferences in between that go, Oh, and I should check the traffic and I should check the weather and check these little things in between. Oh, is he going to be hungry beforehand? Maybe I'll order food for him right before and then having that agent check in with you and say, Would you like me to do this? This is what I've noticed.

I mean, that would be really life changing. And I love the idea that it would help us get off the phones and stop all running around like a like a praying mantis that's always in this weird position. It's crazy, isn't that I mean, I mean, that's the thing and like people are negative about technology. And there's some good points there. Sure. But like, I feel like a lot of the negative people, they have this idealistic view of the world where they kind of assume that we're these serene beings in the garden of Eden, you know, just just gallivanting around and just enjoying, you know, the joys of life 24 seven, but you go out there, you go on a bus, there's 40 people, 40 people staring into their phones, nobody talking to each other.

There's kids on and I'm not saying that, you know, iPads are like, or like phones are the root of or all evil. But like, life is messy. We're not made to live in these cities co located like that. There's a lot of friction. There's a lot of, there's a lot of mess. And the baseline is not perfection. The baseline is life today. The baseline is kids growing up on TikTok and iPad games and computers. And I think a lot of that can be alleviated and improved with removing some of this friction and some of some of this technological smog from our life. So I'm optimistic about that for sure. And also I'll say, and this might be this might be considered a crazy take by some people.

But I think that it's not going to eat up as many jobs as people think, nowhere close to where people are projecting. I think if you even if you look at the results, you know, like at the core, all of this AI technology is going to be using LMS right in the coming years, large language models, the thing that for anybody who doesn't know the thing that powers chat, you put your cloth just token prediction, like, you know, you give it something, it knows the patterns and knows how to complete the sentence. They're good. But Nick, could you run something a hundred times? And if your life depended on it, and be sure that it's consistent a hundred times, it's not even it's not even close.

It's not even close. And so many of these things, so many of these dystopian scenarios kind of rely upon a perfect track record, like how donations are still a massive part of the product. And that's why the main use case today is these co-pilots, it's it's being it's working in tandem with it as a partner. And I don't care how good the cold and how good your optimization is around it. If the thing at its core, which is the LLM still hallucinates one out of 20 times, you're not going to be able to create dependable workers, at least not dependable, more dependable than humans, which, you know, for all our faults, somebody who does their job for decades reliably, like as long as they don't get sick, they get the thing done and they can be held accountable.

What are you going to tell your chat like, oh, you've been there, right? It does something wrong. And you're like, that's wrong. And it's like, oh, my bad, right? Oh, I'm so sorry. You're totally right. And then it does it wrong again. Like as long as we have that as a core, and you know, there might be some technology crazy technological advancement who nobody can predict, but as long as we're anywhere close to the state of these things, and that hasn't, you know, these rates, hallucination rates have improved. But as long as that's a core characteristic, like, I don't think it's just going to be raising jobs like this, because you can't fully rely upon it.

Now, as a few people, you know, I'll go ahead. No, no, I'm just saying like, again, like, there's many nuances to that. And I could dissect my argument and destroyed, like, like going into a certain subsection of jobs and making an argument that that's not going to be the case there. But just generally speaking, people think it's just going to be this magical moment, like, oh, AI is going to work. It's like, man, you can't even write our email reliably 10 out of 10 times. So like, what are we talking about? Yeah, there was a great research paper, and I'm blanking on the professor that did the work.

But it was on the how the hallucinations, which are a negative feature when we experience it are highly tied, like at the base level, to the model's ability to be creative. And the great stuff that it does is tied to one of the worst things that it does, which is the hallucinations. And he was just saying based on that fundamental characteristic without a big technological development, a different way of training these things from the very beginning. He's like, not adjacent. Right. Right. That will be a very big leap, if it's even possible to make it. And one of the things that I notice about the hallucinations is that in domains where I have expertise, I can pick up on those things.

That's a good point. Yeah, it's a great point. What I really worry about are all the times I'm asking you questions about things and working on stuff where I don't have the expertise, because how would I know when it does hallucinate something? I mean, I had a experience recently, I was doing research for a project and it came back with a number of cited articles. And because the research was really important, I was actually going to the sources and looking for specific things in there. One of the citations that was in their MLS citation, there was no way to tell the difference. Actually, didn't hook to a real didn't link to a real article.

And the rest of them were great, but one, it just invented. And I thought, well, how many times does this happen to me when I didn't know what was going on? I didn't check the link because normally I don't do that. Yeah. And there's just this buttress, your point there that like, yeah, humans are absolutely needed for as far down the road as I can see in this process. Yeah. And again, there's certain segments where like, oh, yeah, Igor, but developers, you know, it's like, sure, that's a different conversation and stuff. But like, I'm talking about the overall pattern, I'm talking about society wide, right?

Like, let's not nitpick. And the thing is, it's just, it's just a core feature of it. And I just couldn't help but think, and this is sort of a very, the moment type of thought. But isn't it funny that these models were kind of trained on the corpus of all of human knowledge and all of all the human patterns? And then if you think about humans, like what defines all of us, if you look at every human, every, every religion, every philosophy, every philosopher out there, like at our very core, there's this tension between like good and evil, right, where each one of us is capable of that. But good only exists because there is evil, right?

It doesn't exist in a vacuum. So is it the same thing with AI models? Did it kind of, did it kind of just soak in this character, this balance between, you know, creativity and genius, but also just straight up lying and inventing things on the spot? There might be something there. I'm not qualified enough to kind of dissect that fully, but I think there's something there. I agree with you that there's something there. And I people that regularly listen to the show will know one of the concepts that I am obsessed with and really sparks my interest is that we've got to separate the idea that AI is software as we have grown up with and the way that we know it.

And it looks like software that we have grown up with and spent several decades with. It's hard to tell the difference, right? I mean, Google Maps, yes, it has algorithms doing work there and does amazing things. And so it's easy to think that these AI models are the same thing. And yet when you talk to the people that were in the trenches at the beginning building them, the way they talk about them is they are not programmed, they're grown. And it's a wild concept to think that software could be grown, but they said we're giving it all this training data, we're guiding it in a certain direction, but we don't, we're no one's writing lines of code to say, do this, make these choices, and that that is a real departure from all technology that has come before and comes with all kinds of new things.

Like you said, the light in the dark of human nature kind of being built into it at a fundamental level. And I think that's a really interesting area to explore. It really is. It makes me super curious. It makes me want to read some papers on this. There's got to be something on it. But yeah, as you said, like they, that's a good way of growing. I haven't heard that terminology in relation to it, but that's, that's really good. And all of these scientists agree on another thing, which is like they, there's certain core aspects of all of these models that they don't understand. They call it a, they refer to it as a black box.

Yeah. And whatever goes on in that black box, they're trying to find out, but it kind of, it kind of just knows, it just knows, just predicts it right. Because we showed it the right type of examples, which is, you know, simultaneously fascinating, but also scary. But also at the end of the day, I mean, they're just, you know, word prediction mechanisms. But then if you know, we're from this philosophical arc here, that I, you know, there's an argument to be made that also just humans are words, but like, you know, I don't know, I was born in Eastern Europe, then we moved to Western Europe. I didn't speak a word to German when we moved to Austria.

As a eight year old, I picked it up. I learned how to predict the next words in their patterns. Yeah, just like AI learns how to predict the next word and the patterns of what it saw. It's kind of this, there's all of these crazy parallels between humans and AI and how really the training it with data, there's an argument to be made that, yeah, okay, just predict words, but really like your parents train you with data too. They send you to school to expose you to all these patterns. So you know them in the future, just like we give AI. So anyway, I'll end that there, but that's food for thought.

It is food for thought. I mean, it gets into messy territory and I would just add one thought there, which is that the what separates humans from most of the other species on the planet, some argue is language, which seems to be really baked into us at a fundamental level. I mean, how quickly we pick it up with it's just a natural instinct. I mean, I thought how, you know, I mean, accents are is really just an exposure thing, right? I mean, you take somebody who's born in the in the South and immediately put them in Germany, well, they will grow up with a German accent because we are just absorbing those patterns so well and completely that it just we don't even notice it because it's just so baked into us.

It's crazy. I love that. Maybe on a final note, like all of learning and education is just repeated exposure to certain patterns, right? Yeah. Yeah. The entire school system is just like, you're going to go here and be exposed to this pattern for three years and then you'll know the pattern and then we'll trust you to execute the pattern in the future. It's like, we'll give you a diploma that you're a certified pattern recognizer. Well, right. I mean, well, let's let's dig in there. Do you have any thoughts on how AI is going to reshape education? Because I do think it is making a lot of people think in directions like we just did and then you start to go, what are these patterns that we're teaching in school?

Is it important to show up at a certain place when the bell rings? Is that an important skill and and really reexamining? Well, what is it we should be teaching and how should we be teaching young people at the high school age at the college age? I mean, I think it's an area ripe for a major reinvention in a really positive way. That's that's really easy to agree on. And this is a lot of our cake systems that go all the way back to, you know, the, the, all the factory workers that needed to be educated and then they set up these systems and they still run in a society where the internet is a thing.

So it's kind of, yeah, ripe for disruption. I'm incredibly optimistic for humanity. And, and I think educating people and giving one-on-one attention to people who could have never afforded it, which is because it's like most people, most people can't. It's just, that's just not how the world works. And getting a one-on-one tutor for, for people in, in otherwise rough circumstances is something incredible that I just love so much and AI is already making that possible, right? Just with Chachi Piti, you can have a tutor that is smarter than 99% of humans on any topic already, right? So, so, so I'm very optimistic for that.

I'm very pessimistic for people who are deeply ingrained in the traditional educational system and have this archaic way of thinking of like, oh my God, you had research paper, it's published, it's peer reviewed and goes into here and then it's then it's true. And then we do it this way. I understand the values of that. I'm not, I'm not saying that there is, that's, that's, you know, a system that serves our purpose. But in this day and age where you can get, get personalized information delivery at scale to essentially everybody, I think there is going to be, I'm actually surprised that it's not emerging already.

There's going to be a new and better model where it's a mix of human and, and technology, human and AI, where, where maybe the human creates environment that the experience and the connection and makes you learn the social skills and, and showing up at the bell. But then the AI is there when you're sitting at home doing the, the homework and tutoring you one on one for every single assignment in a language that resonates with you with examples that are taken from your life, from your experience in the past weeks, with, with an authority that, that no human can match because it understands you and it knows that, hey, if you, if I'm going to push harder here, he's only going to retreat.

So I'm, I'm going to take the more elegant approach because I've seen this a million times before. And I think that level of attention, only some of the best teachers can bring to the table and, and even them like, cannot deliver that consistently and certainly not at scale. And I think AI is going to make that possible something incredibly, incredibly optimistic about that. And I, I hope, and I can't wait for my children to be educated in a way that is just just, just personalized, but always in combination with humans. Like I, I would, you know, I think it's that mix that's magical. It's not about replacing.

It's about enhancing. And yeah, it's just a better world. I mean, that's just so clear to me. This episode of the next daily show is brought to you by Zapier. If you've ever felt buried in repetitive work, copying data, moving files, sending follow ups, you know, it's like death by a thousand mouse clicks. Zapier has always been the tool that fixes that connects over 8,000 apps Google Drive, Slack, Notion, Gmail, MySpace, you name it, so your tools can finally play nice together. But here's the big shift. Zapier now lets you create AI agents with their chat GPT integration. Think of them as tireless teammates who never complain, never take lunch, never get bored of doing the boring stuff.

I've actually made several of these little agents myself. No Python, no JavaScript, just pure vibe coding, which is my favorite kind of coding, because even though I have no idea how to code, I just talked to the AI, tell it what I want, and almost magically works. For example, when a podcast guest books a time, Zapier can send them a personalized email from me, preparing them for the show, create a draft of show notes, update my calendar, and even prep a social media post automatically, all of which freeze me up to focus on the important stuff, like taking credit for the amazing work my agents did.

And now, Zapier just took it up another level for developers. Zapier MCP is available right inside chat GPT, which means you can connect those 8000 plus apps and trigger workflows just by writing what you want to happen. Literally, you tell chat GPT, send this file to my team in Slack, update the spreadsheet, and draft an email, and Zapier plus chat GPT, figure out the right tools, do it for you. Getting started is simple. Head to the Zapier chat GPT MCP server, and add the tools you want chat GPT to access. Follow the steps in the connect tab. And if you're an admin on a chat GPT enterprise account, you can enable MCP across your entire workspace.

So if you're ready to stop wasting time on busy work, join the AI revolution, and make a little automation magic of your own. Try the Zapier chat GPT integration using the link below. I couldn't agree more. I mean, I know from running a company that has helped students raise their SAT scores for 20 years that one on one is the unlock. And it's not about necessarily just raising that SAT score. When a kid has one on one attention, suddenly, yes, their SAT scores will improve. But it is their entire model of learning is transformed when they have that one on one attention, suddenly their grades improve.

Suddenly, they are excited to learn about new things. And that only comes from that one on one attention and focusing on that kid. What are your goals? Where is it you want to go? How do we connect the things you have to do now to these future goals? And that's very difficult to do with 40, if not impossible, with 40 kids at once. But one on one, it's made for that. And even on the not replacing humans, though, I think it's the perfect example where, yes, a lot of the drudgery stuff around teaching will continually be handled by AI. It's great because if you have a thought in the middle of the night and your teacher's not available, you've got someone you can get some answers from and some help from.

And yet, at the same time, there's that human element that can never be replaced. And I, there was a recent example when I saw one of our teachers working with a student, and there was no actual language involved in the in this moment with this kid where the teacher's talking to the kid and asked if he had done his homework, and he said, yes. And the teacher picked up on something in that, yes, that it didn't seem 100% truthful. And all he did was just cock his head to the side and give a little look with his squint a little bit. And this kid broke open and said, okay, I didn't really do it, but blah, blah, blah, blah.

I have these reasons. I did, you know, a little piece of it. I tried. And I thought, well, that right there, right there, that moment is why it's all about the technology meeting the human. And that's where the further unlocks will be in learning and education, it's it's the meeting of these two things. And so everybody worrying about being replaced is not I don't think it's the right framework. I don't think that every job's gonna be put into, oh, it's either AI driven or it's human driven. It is all about enhancement of the human skills and amplifying that person. I mean, I get it though, because everybody wants to simplify it, right?

It's so much. It's such a whole new domain. Like, it's like we've discovered a new continent and now we don't know exactly how things function like continent, but it's a good continent. And we want a lot of stuff from the continent or like we have all these visions. But like, it's not that black or black and white, but people like to think in those terms, because if you're like, Oh, continent good. Okay, okay, continent bad, continent bad, you know, like, like, it's not that simple. A lot of it is gonna be hybrid. A lot of it is gonna be a fantastic teacher. Or maybe maybe let's say an average teacher who would really struggle with holding the attention of maybe his exceptional in certain domains, but maybe he struggles with the holding an attention of 40 40 kids that I don't know, maybe are 16 years old and particularly difficult.

And all of a sudden he can kind of initiate an exercise, maybe lecture for 10 minutes, but then go into a piece where everybody works with the AI for the next 15 minutes to develop an idea. And then somebody comes up and presents that. And then the the AI's work together to come with a conclusion to distribute the learnings of all 40 kids to all of them by synthesizing it and recontextualizing it to the other kids. Like, it's just stuff that a human couldn't do. But all of a sudden in the classroom, like it's like, yeah, kids get out your iPad and let's do this. And the results are because learning at the end of the day, it's so much fun.

It's one of it's one of my favorite activities on this planet. And it hasn't always been that way. I was I was average student that was just getting by with doing no work in school and just sort of like, Oh, make you know, I want to be done with this so I can go do what I want to do. But now in my 30s, I'm like, there's I can't imagine anything more fulfilling than learning a new skill every time I get a little, you know, too confident or good at one thing, I find a new thing to suck at again. And it's wonderful. And back and there we go. I'm back at Squares a year old. And I just just I just love learning.

And I think we're going to be able to unlock that enthusiasm and excitement and passion for other for kids who with the right mentorship, it's always there. And the right mentorship could bring it forth. My hope is, and my belief is that AI is going to make that happen and make it possible. And we're not we're not years away. It's already here. We just need to we just need to massage and the details of it. Yeah. So if you are, let's say, any age, and you're curious about all this, you're listening to this conversation, you go, Well, what should I be focusing on next? I mean, is it about taste? Is it getting involved with the coding?

Is it about being more entrepreneurial and trying things and experimenting? I mean, where would you guide someone in where they should direct their curiosity in this area? So in my experience, it really, it really depends on where you are in life. And I'm going to try and bucket this into two buckets. Generally speaking, you could think of this as people kind of in the first half of their life or the second half of their life. But honestly, just as I said, if you're picking up a new skill, you're in the first half of that skill all over again, I don't care if you're 80 or whatever, right? Like if you're if you're in a place where you're not sure what the direction is, go out there and just I don't know, get out of your door, turn to the right and just walk for two hours and experience life.

Like it doesn't matter. Just just go collect experiences, go out there, move, right? And by collecting new experiences, you're going to learn new things, not just about the world, but mainly about yourself and how you interact with that, what you like, which don't like develop your taste that way. Say yes to things where you're not sure right away. There's a season in life to, you know, put boundaries and say no to things. But if you're not sure about your direction saying yes to more things is the way to discover that direction. Now, if you have that direction, I would recommend finding ways to use AI to go faster in that direction.

So for a lot of people who we work with a lot of people who are, for example, in their 40s and they have a thriving career and they've been doing so well, but out now all of a sudden AI is a thing and they got to learn this thing and it's going to impact their job. But they're like not a tech person. And they've not like interacted with computers very well over the past decade. They maybe sort of even kind of ignored it a little bit. They were like, this is not for me, but they're killing it at their job. Those people need to figure out how to go faster and how to unlock the capabilities of AI for what they're doing.

So they can increase their output. They can increase their their, they can get time back by using AI and they can increase the leverage that they already have and make the lever even bigger. So those are those are two different things. For people who are in that second camp, it really is about depth of context. The context is still already you, you're probably kind of clear on your goals and what you're doing and how you're doing it and how you communicate and what your preferences are. You need to formalize that and make sure to start working with AI at a level of well, more self knowledge that you communicate to it, more intimacy, more tell it more about your context and make sure you do that.

And then even some of the peripheral applications like meeting note takers start working better because if it has context on you, it will summarize the whole thing into the front way. Everything just starts unlocking. For the people who are just starting out, I will just recommend getting hours. And honestly, you can kind of do this fun exercise, which is one is using it daily. But the second one, which I like is if I, I work with to do lists, I love to organize kind of my thoughts and things that I need to do to write them down. If you make a to do list for all the stuff that you're going to do over the next month or want to do, find a way to use AI for every single one of those items.

I don't care if it's like picking up your laundry and the result is not going to be that phenomenal. Just find a way. Hey, how could you help me now? Now, I'm answering the question of like, okay, you go, how do I do that? Literally open it up and be like, how could you help me with the task of picking up my laundry? Enter. It's going to tell you something. It's going to be like, oh, I could look at Google map routes and optimize your way there. It's like, really, you could do that? Okay, go do that. I could find the 10 closest laundry and compare the pricing. Okay, okay, good. Go do that. That's a good idea.

And it's that's a trivial thing. I'm not even talking about big decisions or relationships. Relationships. Oh my God, it's so good. You just need to ask it and you need to lean into that curiosity rather than leaning out because of some underlying fear of anxiety or uncertainty. You can't, you know, it's a choice. If you make it conscious, it's a choice. So just do that. Just lean in, be more curious and yeah, the path is going to reveal itself. And if you have doubts, type them out and hit enter. It's a chat interface. I always say, it's not one message GPT. It's chat GPT. So if you have worries or criticism, vocalize it whether for voice input and put it into chat and hit enter or type it in your keyboard and hit enter and see what it says, get fluent with it and build a relationship with it and you'll thrive.

Yeah, and it is amazing how sometimes those simple things can end up opening your mind to what is possible. One example that pops into my head is take a picture of what is in your refrigerator and send it to your model of choice and just say, I want to make something for dinner. This is what I've got. And you're not even telling it what the items are. It's just going to pull that out of the picture. I'm not a cook, a chef by any, any means whatsoever. I am amazed at what it can come up with and walk me through the process of just making a meal on what is available in the refrigerator. And someone mentioned that to me a long time ago.

And it opened up my mind on like, whoa, there's a lot more possible here than just asking it about some random textbook and got me on this on this journey and this path that I'm deeply on now. I love it. And then I would just add to that, Nick, like what you said there is one of the best examples ever. But what you're describing, if you abstract away from that, is you give it a task, but then also you give it context or you could think of it almost as data, right? The picture is data. And it's like a big word. People are like, oh my God, data. What do you mean? Like a picture is data. A picture says more than a thousand words, right?

So if you combine a picture with a simple task, you get magic. In the same way, you can get data from all over the internet, all of the internet is data, right? What I say, what I say as a good example is if you go into a YouTube video on a topic that interests you, there's a comment section with hundreds of curious souls asking the most burning questions on their mind. You can literally just drag all over those comments, copy them, put them into chat and then work with that. If that video is relating to something you're working on, I'm sure there's some signal in there. For example, you know, me as a content creator, I do that all the time.

I find ideas from new videos, from comment sections of other videos, like people are asking the questions already, or you could take customer support data and analyze that for the recurring patterns, whatever you is. And when in doubt, when you have any sort of data and you're not sure what to do with it, Igor told me to look at data. And apparently that's supposed to be helpful. I don't know what to do with it though. Ask this question. Just say, what patterns do you see within this? And then add whatever. You can add your refrigerator picture. It's going to probably see some recipes in there without you even asking just what patterns do you see and then add a thing and, and you know, go have fun.

Yeah, the one other example that pops to mind is that this was just a few weeks ago. And this was with GPT. And I sent it stills from a recent podcast episode and said, what do you think about the lighting and the setup in this studio? And then here are some other stills from other channels that I really like what they have going on. And I want to lean towards this direction. And it was really wild. It started asking me for pictures of what else is in the studio. Can I see your lights and can I can you you know, try to get 360 degrees in here? Can you take pictures of the cameras? Okay, sure. Started doing all of that.

And it said, okay, here's my analysis. Your background is really well lit, but it actually draws it's so well lit. It's drawing attention away from the speaker, especially if you're doing a solo episode. And then it said, I got so specific where I'm excited. These big key light over here and then a fill light on this side. And I said, I need you to turn down the fill light by 10%. And then send me a picture of that. And I'd send that as we got to go another 5%. And then it said, can you add a hair light? So it said, you need to get something, put it really up high and let's separate you from the background.

And sure enough, very first video I put up with this, there are comments in there that like, Oh, the set looks really nice. And I was just blown away by like, it wasn't just me going like, Oh, I agree with you. It was real humans notice the changes and said, wow, that looks really cool. And they didn't know what they were picking up on. But I thought, what a wild experience that is that is that is I think I got to say your background is gorgeous. You're lighting everything. It's so good. I just I almost feel embarrassed. I moved to a new place a month ago, and I traveled for most of the time. I'm gonna work on my I'm gonna work with chat to do exactly what you do.

I need a hair light up into the up in here. Yeah, well, luckily, you're so animated with your facial expressions and there's there's nothing distracting behind you. It's true. It's like it's silver linings. I love it. It still works. It's so good. Or if people want to know more about your school and your offerings and your trainings and where to find you online, where do they go for all that good stuff? Oh, it's simple. So look, like we have the club where for a dollar, you can get 30 days of progression through this map through these levels and I'll guide you through it. And then you can make a decision if that's worth your time to stick around for the community, you can find all that under AIadvantage.com.

It's the club that I'm talking about the AI advantage club. And then me personally, I run a YouTube channel. I have an Instagram page, you can just find AI advantage on YouTube. Every week I look at the new releases and try to filter for non-technical humans, consumers who want to figure out the how to use this stuff, how to apply it to their life. And then I just have a personal Instagram at Igor Pogany that's i-g-o-r-p-o-g-a-n-y. It's a Russian name with a Hungarian surname, but I'm Slovakian that grew up in Austria and lives in Portugal and works in the US. It's a mess. Don't ask me why. It's just that.

And I would love to, to, you know, see you and teach you and all of that. It's what I do on the internet. I love it. And we'll put links to all of that in the description so that if anyone has trouble spelling that they can easily find it and they'll need it. Yeah. I know how it goes with my last name as well, which is Stanley, but it spelled S-T-A-N-D-L-E-A. And so the story goes, there was someone many generations ago in the early days of the United States who was trying to evade the law. And so it was spelled S-T-A-N-L-E-Y and changed the spelling after doing something bad. We've not been able to track down what that actual thing was, but that is the family story on how we ended up with this very unique spelling of my last name.

Oh, really? That is so interesting. You have a bad guy in your ancestry, so that means there's a little bit of that edge inside of UNIC. The light in the dark. We all have it, but with you it's official. That's right. That's right. The light in the dark. Well, Igor, this has been such a pleasure and we've gone in so many interesting directions, but there are so many just practical applications and tools that you've mentioned today. I really enjoyed the conversation and I can't wait to tune back into your channel and see what comes next. And as a side note, just as a closing side note, because I think everyone will find this interesting too.

It was a very simulation-style moment where I was listening to your content on Spotify. That's where I usually consume podcasts. Really liked what I was hearing and listening to different interviews with you and the things you were bringing up. And then someone from your PR team reaches out almost right afterwards and says, hey, this seems like this would be a really good fit. Let's have Igor come on your show and I just thought, wow, that's the magic of the internet, of these algorithms. Yeah, it's just a just a neat little serendipitous moment. And so wanted to share that with you that I was I was consuming your work before before this moment and it was very appreciative of it.

So really grateful for the conversation. Oh my God, what an honor and what a conversation this was. I enjoyed myself so much to make like we went all the way to the highest levels of philosophy, but also dipped down into conical multiple concrete prompts for people to run right now. So if you don't know what to do next and that's on you, not on us. This was great and it was it was an honor to be on here. Nick, thank you so much. Absolutely. All right. I'm going to see you soon, Igor. All right. See you soon. Okay, everybody until next time, ask questions. Don't accept the status quo and be curious.