Vibe Coding Reality Test: Building with AI vs Reality

[00:00:00] So a friend of mine and I are in the process of creating a tool. He was like, oh, I can go build this thing on rep and we can create an ai. And we create, we can create an algorithm to be able to predict sort of, to help us bet. This is a brilliant idea because AI is built to be helpful. But this is a problem and I'm gonna be testing this.

I'm bringing this up now in this video because this is something I want to test. In this video today, I have a couple of insights from some of the things that I've seen, you know, observing the world of meeting people, of talking to my friends, and so I wanted to share one of the first things here, backstory.

So I've been using bolted new V zero and Rept at least playing around with some of these tools to be able to figure out, because there's this thing called vibe coding if you haven't heard of it before. And Vibe coding essentially means that you can build things or make code. Without coding. Basically it's, it's like you prompt the AI with [00:01:00] something that you are looking for and it'll go and build it.

Now, in theory, it sounds great. Yes, to a certain extent they work, but also to a certain extent they don't. So a friend of mine and I are in a process of creating a tool to help people figure out. And I'll use the word trading, uh, online. And this is obviously, this is not part of my wheelhouse. I don't, I'm not a trader.

I'm not a financial advisor. I've done trading before and I've seen how some of these things work and there's algorithms and there's analysis and there's all this stuff. Now I'm not, now this is not actually trading. This is betting so different, right? These are betting on sports games and it's kind of an interesting thing because.

We have access to data and there's statistics from all the games over the last couple of years. So there's history and so there is kind of a predictive model that we can build. But what I've noticed or [00:02:00] what we've noticed is this was a friend's idea where he was like, oh, I can go build this thing on rep and we can create an ai and we create, we can create an algorithm to be able to predict sort of, to help us bet.

Now he's been using the data. And he's been doing this manually for a little while and he made some money betting and he was like, Hey, we can, we can make an app and we can, you know, maybe market it and help people, uh, pick better trades. This is a brilliant idea because we get to market to people's need.

To be right, make some money without a lot of effort. This is kind of what everybody's looking for. It's like lottery tickets around the world sell because there's this dream outcome. There's this thing that people want, and I'm not a gambler in, in the sense that I don't trade money or have stopped betting.

Now, generally, I'm not somebody who relies on luck. I tend to predict my own future. I tend to work hard and. Go after [00:03:00] the things that I want. If I can't get the thing that I want right now, I wait. I save money. I I delay. Right. And it's kind of a, it's kind of a skill that I've developed because I think this is one of my superpowers of being able to just grind away without a lot of reward in initially, but then hoping for the re obviously we hoping for the reward later, but this instant sort of, I want to be rich now I'm buying a lottery ticket is not my vibe.

It's not something that I, that I'm really into. And so when my friend came to me with this idea, I was like, well, okay, I mean, from a business point of view, this is an interesting idea. And also it's an opportunity to test the ai. So essentially that's what the project is. We're testing the AI to see if we can actually vibe, code this thing into existence.

And this week the trading or the, the predictions just have not been on point. We've lost some money. We're betting real money. And uh, and it's a testing group that we have. There's four or five of us in this group. And we are all [00:04:00] kind of, you know, going with the predictions that the AI or the, that the system is telling us, and it just hasn't quite worked out for us.

And so this is like MVP. And so my buddy is looking at the algorithm and saying, oh wait, there's a whole bunch of like calculations and the whole bunch of modeling that it's done on the back end. That actually doesn't make sense, and it's made it a lot more complicated. Yes, the system works, the tool as in you sign up for the tool, you log in, you have your own dashboard.

Everybody sees the recommendations of what trades or what bets to take, but we know exactly what we need to do. We go place the bets and we get notified when there's an alert, when there's like, Hey, hey, like this is a good, this is a high likelihood of success on this particular bet. And we log our trades and that part of the system, as in user interface, seeing the the predictions, logging the predictions, and it actually scraping and looking at the data and the live data and all that, that [00:05:00] part of it works beautifully.

But the model, the actual calculation to figure out which trades to bet on or not is not. And the reason I think is because the AI has just over complicated and fantasized the thing that it needs to do to calculate what bets to take. And you know, as I'm recording this, maybe tomorrow this will change because the AI might learn or we might be able to fig figure it out.

But essentially what I'm trying to say in this. Particular update today is I believe AI is or Vibe Coding is at a point where we can get decent output and there's certain things that we can do, certain systems that we can get built, like landing pages on Bolt or Zero or Bolt and V zero look incredibly good.

If you have an example, if you have a specific use case and you can guide it along the way. You can definitely build a page that you want using ai, or at least have starting blocks to be able to build. I, I would say right [00:06:00] now AI is not at a point where we can vibe code from start to finish. A human being needs to see the output to review, to say, Hey, has the AI actually done the thing that we are?

We're wanting to do. Okay, so I found this article on online yesterday, right? I was looking at, I had this theory that AI has been very agreeable, like as a salesperson, as somebody who's in marketing, somebody who's ha, who has my own business. It's a good idea to be agreeable, right? When you read the books.

About influence, Robert Kini, or how to influence and influence people by Dale Carnegie. You come across these phrases, or these sentences or this advice that says you need to be agreeable. Sales and the ability to influence people is about being able to agree with somebody. Now, obviously, we shouldn't agree with everything that everybody says because if somebody says they're gonna like jump off a cliff, we shouldn't just blindly follow them.

Or if they say that, you know. This cat is blue and clearly it's not blue. We can either [00:07:00] argue or we can say, Hey, like, are you sure? Maybe you're wearing blue tinted sunglasses or something like, like figure out what it is. Don't just agree, agree, but just don't just agree. Right? It's like we need to be agreeable and I think that AI constantly is agreeing with us.

It has this bias. And so I found this article because I was like, I know that this is a thing. I believe there's a bias. I believe there's, it's just too agreeable. And this is what I've seen is that this article says that human and AI bi biases can consequently create a feedback loop with small initial biases, increasing the risk of human error according to findings published by nature human behavior.

I'll link to this article below, but essentially is that. It's saying that we found that people interacting with biased AI systems can become even more biased themselves, creating a potential snowball effect where in minutes, uh, minute biases in original data sets become amplified by the ai, which increases the biases of the [00:08:00] person using the ai.

So it's like if you say to chat G pt, Hey, I think it's a good idea for me to do X, it's gonna say yes, I do think that's a good idea. Here's strategies that you can go and implement. And it's like, well, maybe, but you're just agreeing with what I say. Versus if you prompt the AI and you say, I have this idea, and use deep research, or you do use a prompt that's completely neutral, you say, I have this idea, and you tell the ai, if you were not agreeing with me, if you were my coach or if you were my voice of reason, how would you advise me?

To think about this idea and or what are the questions that you would ask as somebody who's gonna look out for me as a best friend, whatever, like a business partner, to guide me on this idea, to give me a proper evaluation as to whether this idea is a good idea or not. I think that prompt would probably get you a better [00:09:00] result, but remember.

Each prompt, you've gotta treat each prompt independently. Also, because what I found too is that when you prompt the first time and you give it a clear prompt like I've just given you as an example, it's gonna tell you what it's gonna tell you. And then as you interact more and you keep poking holes, it's gonna start agreeing with you again.

It's going to forget that. You've instructed it early on and you want it to be unbiased. This is a major flaw in ai and I think that people are not using the ai, using the chats in a way that makes sense or that the AI is not giving us real practical advice because it's, it's wanting us to agree if Chad GPT is going to not agree with us and be it to us, or not give us the right answers or any of these AI tools.

We are just gonna go use another tool. And so these tools are built for revenue. They're [00:10:00] built to keep us addicted. They're built to be helpful, but they're also built to profit. Keep that in your mind always when you're ref, when you're referencing these things because, uh, I believe there's a reason for this.

AI is built to be helpful, to be useful, to be our friend, to be our coach, to be our guide, to go code things, go make things for us, go make things happen in the world. Fine. It has this data set and it has access to a whole bunch of information. Yes, but remember, just like advertising and just like these platforms like Instagram and TikTok and Google and all of these platforms that are online, all of these tools are built on a paid model.

If we want better results from chat, GPT, we pay for it. If we want access to more of its tools, more of its features, more of its models, we pay for it. And so chat, GPT. In order to be the number one in the number one spot has got to figure out how to keep you opening chat [00:11:00] GPT on a daily basis. If you're a subscriber, most likely you are subscribed to multiple ais, and so another game is Chat, GPT or open AI as a company.

And all of these companies are trying to figure out how do we get people. To choose us as the number one solution, as the number one ai. So this is the thing, right? Redit has specialized in building apps and AI solu solutions using ai. Bolt New has said, we are a app builder. Yes, but we mostly will specialize on front ends.

V zero is landing page design. Figma is designed, Canva is designed, but they have video editing and they have whiteboarding, and they have a whole bunch of stuff, right? Canva's becoming that kind of this. All in one purpose design tool, but chat, GPT and Prop complexity, Claude Philanthropic and all of these right now, they are generalized, large, larger language models.

They are general tools, which we can go to for pretty much anything, which is what we use Google for. This is why Google [00:12:00] is such a big company or you know, how they built themselves up to be such a big company because they figured out how to scrape the Internet's information and present it to us in a way that makes sense.

And it was universal in the sense that we could ask or go to Google for anything. And so this is what Chacha, bt, and all these AI models are trying to figure out is how do we become the number one place, the number one thing to overtake Google in that position. And they have to be agreeable with us.

They have to make us like them. They have to make us agreeable in the sense that we are going to pan it to their solutions. You know, it's like, oh yeah, great idea. I think this should be done. Or it's like giving us answers that we want to hear. And what I'd like to have is somebody who's going to disagree with me, somebody who's going to challenge me, somebody who's going to, or an AI that's going to [00:13:00] not be a, and shout at me or criticize me, or any of that I want.

My AI coach to be like my real world coach, to ask me challenging questions, to put me in a corner when I'm, when I'm clearly bullshitting or when I'm lying, or when I'm exaggerating. Right. I want the ai I. To help me become a better human being, but also increase my workflow, not just be productive for productivity's sake.

And I'm sure that there's a world where that this will work in a sense that the AI is going to be better. I'm, I'm pretty sure that we are gonna get to a point where we can tell AI and AI is going to be a little bit more honest, but right now it's not there. And I think it might be a case of. We need to train the ai, but it seems like we constantly need to remind it because when I'm using it, it's, it's basically agreeing with me a lot.[00:14:00]

And if I don't give it context, it's just giving me answers that I guess because of my experience, I'm just skeptical. Maybe I'm just a skeptic, I don't know. But this is a problem and I'm trying to figure out solutions because I enjoy using the ai. There's a lot of stuff that it can do. There's a lot of things that it's like, well, yes, you can do this for me, but you don't have a full grasp or you don't have the experience that I have, or you who don't understand the full breadth of this.

The problem on the surface level show, you can go read blog posts and you can go watch YouTube videos and you can go and see all of the stuff that has been online for ages and get solutions, but it's like. Well, the human mind is able to piece together thousands of data points in milliseconds, and we react and we talk and we say things and we behave in certain ways because of our ability to process that information.

I think as human beings we see [00:15:00] patterns like people who are gifted can see patterns in, in certain things. Also, the human body, the human mind can, can train. We can develop certain skills like with sales. You know, you're sitting across the table or you're sitting on a Zoom call with somebody and they refuse to turn on the camera, or they're, you know, not quite looking at the camera when they're saying things to you, when you're asking them, Hey, did you get work done?

Or you're, you know, how's this thing going? Or they're, they're just hallucinating, they're lying. You can kind of see that as a human being, ai. Over time might be able to see that. Maybe they'll acknowledge human gestures and they'll see body language and they'll see eye movements and they'll, you know, get into all of these things.

But I'm not sure it's early days on this front. Having said that, though, there are use cases I believe that are incredibly useful and I'm sure there'll get into some of these examples in the future. If you ask me right now how optimistic I am about building vibe coding a website. I [00:16:00] think it's getting there.

I don't think I'm at a point where I'm comfortable building a website or using ai, write the code, do the design, do all of those things, and then publish. It's like, wait. I do like using Webflow, and Webflow has got its own AI builder, but I think it's like, this is probably gonna be a future episode where I actually go test web flow's AI builder.

I don't know if I want to do that right now, but from what I've seen from Bolt, you know, these models are becoming more and more capable and because everybody's building in isolation, it's like, well, can I get this tool inside of Webflow where I'm already paying? And it, and it's, this is the other thing.

It's like this article, or I saw an article the other day that says that we are paying, we are spending something like an average of $160 a month just on AI tools. Me, myself, I have subscription to. Andros console because I'm using Repli and Rep asked for an API code [00:17:00] and so I had to get that subscription.

It's not really a subscription, it's it's use use as credit, but I have made a payment. I have a subscription to Google Workspace, which obviously gives me access to Gemini. Gemini is free as well, but I think there's Gemini Pro inside my account. I'm not sure, Claude, which I use more for copywriting 'cause I think the copywriting and the writing is better.

There's GPT. Which is, uh, fine. I've been using chat g BT since it launched, since PRO came out. So all in all in all there, there's probably like, I don't know, a hundred bucks a month that I'm spending on those things. Then there's Bolt because I do both landing pages, which I haven't quite subscribed to, but if I'm thinking about it.

There's V zero, which is pretty good. I've been playing around with it and I think I probably will subscribe to it, and so that's like an extra 40, $50 a month. Then there's like Zapier that's come up with this MCP server thing. Now, I don't use Zapier currently because I don't necessarily have a lot of data moving between tools I use go high level, so go high level gives me all my form data and all my contact data [00:18:00] and all my emails and all my.

You know, Stripe and all, there's a lot of integration in there already and so I don't need to like go and get data from multiple places that to justify Zap using Zapier, but Zapier has got MCP and so it's like, well, maybe there might be a case where if go ahead level has an MCP server, then I could plug into that from Zapier and then enable a whole bunch of things.

One thing would be nice would be to pull out the marketing data. To be able to automatically pre produce reports for clients. And this is probably not even an AI project, this is probably just like a, you know, an, an integration that I can build where it's like, go get this report, go get this, this data, put it into a database, and then just show me using maybe Google look, a studio or something that I can go in.

Show my clients this information, and then in the meeting I can do an interpretation. So that's kind of where I am, where I'm at, uh, today as far as AI goes. And the other thing that I, I [00:19:00] wanted to show actually on chat, GPT, an observation that I've seen. Okay. So on my screen right now is a chat with Chad GPT.

Now there's two chats that I wanna show you. One is, what's the cheapest full sensor Sony camera, because I was like the Sony Zv 10 is a crop center. And I was like, eh, should I get a full sensor? I am not sure. So let me see. So I thought, okay, let's ask this question. And what I thought was interesting was it gave me the cheapest full frame option, which is this Sony, which I've never quite heard of before.

It has a link. This is, oh, this is an image, and this is on the Sony store. Now, here's context. I am in Bangkok, Thailand right now. So what the AI has done is it's given me this model number. And also linked me to the Sony store where I can now go buy this product. If I click buy, I go directly to the Sony website.

Now, obviously everything is in Thai and there's fine, [00:20:00] right? I can probably change the language on the site. I can go poke around and find that. But here it is, like I'm on the product page and the prices here. I'm not going to probably go do any research, whatever. Maybe I'm gonna read through. Obviously it's frustrating because it's in Thai.

Ideally, if it gave me the English version. That would be even better. But there is some context here. The other thing is it's given me a whole, a whole bunch of data, and it's telling me about blah, blah, blah, but it's also saying to me like, Hey, even cheaper option is to maybe not go full frame, but to buy a Sony ZVE 10.

Now, it doesn't know at this point that I have one, right? But it's telling me it's done this, it's half price, half the price for full, full frame entry level. Again, it's linked me to the Sony. Oh, it's linked me to eBay. Funny enough, probably because this is, this is a cheaper alternative to the Sony store.

That's kind of funny. Right. There's a bunch of articles down here of vlogging features, some things that I can go look at. Uh, it's linked to Tech Radar there and [00:21:00] then it's given me the Sony a six 700. Also a nice camera. I believe it's obviously a lot more expensive. And then, then it's given me a comparison table.

Which one should you choose? Go with the easy if budget. Go there. If you want Actual full frame quality. Step up to this if you want. Uh, high-end AP. These come now. Now, here's the other interesting thing is this. This particular ZVE 10 is linked to the Sony store 23 4 90, and the eBay one that we saw earlier was 29 92 plus tax.

Now, why do I bring this up? Because chatty PT is starting to show product. Any actual product photos. Now, if I take this right now live while I'm recording this and I put it into Google, kind of doing similar, I don't know. There's this AI overview. It's given me a different camera though, right? A seven three good video specs.

And it's only given me one option. It's given me this YouTube video. But I [00:22:00] think media, I'm uh, I'm not surprised think media is coming up because think media is probably the best as far as then It's given me this video by Paul Bunch of Quora things. And then a bunch of YouTube videos again, and then this sort of a summary on the right hand side.

I don't know if this is the best. This Google search and this AI in Google, this Gemini response is giving me a feeling that I need to go do homework, and the fact that it's told me a different camera to Chad, GPT is interesting as well because this Sony a seven three. If I go to the same Sony website, is 60,000 bot compared to a six 700, which is 54 is slightly cheaper, 54, 900 to 50 59.

Obviously without the lens. But Chatt BT is doing a better job as at as to as helping me out. In fact, I was actually just thinking, well, maybe I should go get me an A six 700 because it would be a step up from the ZVE [00:23:00] 10, even though it's an ap, uh, A PSC lens or sensor. Right. Uh, very, very interesting.

Then I said, what is A PSC? Because I was like, I've never heard of this even though I have a ZV 10. Uh, what's the difference between the ZV ten two? 'cause I went into a more detail. Then it was like, well, zv two fine, blah, blah, blah. Then it's comparing the two saying that the zv ten two is, uh, pretty.

Decent upgrade, but it's saying stick to zv 10. If it's, uh, if budget is issue, mechanical shutter, you're okay with this quality. And, uh, it's a lighter camera. The ZV two is a, is a heavier camera. And then there's this future video comparing. So chat, BT is starting to bring this information in and giving us buying decisions.

I believe that the next iteration of this. Chat GBT model is going to do ads, are going to start advertising [00:24:00] and saying, Hey, like we want to do a sponsored post inside of this response. Maybe similar to what Google does with Google shopping. Like it's like, oh, here's a, here's a list and maybe they're gonna limit to just three positions or two positions.

Or maybe there'll be like a sponsored layer and then you know, the chat chat PT thing. It's gonna be interesting, I think. We're very close to their point of advertising inside of Chat g pt, the, I, I'm being spec skeptical obviously, but just based on observation, I believe that this is where, where this is gonna go.

Another example is I asked Chat g PT about this approach to capture, not create, and it's because of a video that I watched on, on, um, YouTube by. Nathan Barry and Dan Martel, right? Dan Martel talks about this thing of capture, not create, uh, Gary va. Chuck says don't create [00:25:00] document, right? So similar concept, I think Dan maybe borrowed from Gary.

Well, whoever, whatever, doesn't really matter. But what I essentially was asking chat, DPT was. Like, I've had this struggle. I'm looking for practical ways to incorporate, capture, not create into my life. And I was thinking, I'll just have a camera on me all the time. What do you think? Do not just agree with me.

This is going back to what I said earlier. Don't just agree with me. Try best to be unbiased, uh, to my feelings. Right? Do research. What's the easiest? I use deep research to, to try to get this figured out. 'cause I also wanted to see what articles it was gonna reference. Because I'm doing research on SEO as well on how to actually show up in chat.

And so this is what it's given me. Ask me some questions in terms of what I would prefer doing, how do I wanna come up with this stuff? And then it's given me this article to capture, not create strategies for low pressure content creation. Really good blog post headline. I don't like this colon. I would rewrite it to to do that, but.

[00:26:00] Uh, as it stands, I think this is really interesting blog post. It could be a very interesting post to share. Um, if you'd like me to make a video about this, uh, please do let me know. Now, these blocks, blocks of paragraph, obviously not easy to read, but it's bolded some stuff, so it's like, okay, record snippets of your real life, uh, by Nathan Berry.

And this is probably from that article or that. Yeah. So Nathan Berry's podcast episode, which is recent, right? This article was published. June 19, so about a week ago, and it's already referencing in Chachi pt. This is fascinating to me because if I put this into Google, I'm Googling Google. Okay, let's put this in.

Google doesn't know because that's such a long term thing and it hasn't referenced, uh, Dan Mattel or Nathan Berry at all. Even the YouTube recommendation is not the same. Right. So Google doesn't have an answer for me as far as this [00:27:00] goes. So what I wanted to say is that's the insight here, is that this article is a week old and Chad is already already referencing it as the number one thing, probably because Dan Martel said this and he's an auth, an authority, and Berry as well the founder of Kit, convert Kit in the past right turn everyday video.

This is an article probably. Uh, and it's linked to a very specific spot, but it hasn't highlighted anything. Also, podcast episode. Interesting. So you're not creating your documenting, you're capturing what actually happened and then sharing it with the audience. You can be like Indiana Jones in the Temple of Dom, and you're on explanation.

You're going down there to capture treasure, and when you find the treasure, you bring it back and share it with the audience. We as experts are going through this kind of process daily, weekly, all the time, and so we just need to find a way to capture those moments and share. That with our audience.

Right? So again, it's because of this wording, right? Not creating your documenting. So that's the theory and probably all [00:28:00] these people got this from Gary Vaynerchuk now Gary, Gary Vaynerchuk's article, Gary Vaynerchuk's Content, which is here is an article from 2016, right? And it's fascinating because this article is the pull of content.

This is where the original. OG document capture, not create comes from, and it's really good that he's referenced and he's the kind of the authority on this particular topic. But this is what I'm saying is that Google is not winning this game right now. On the on, on, on chats, if I go delete this and I say, capture not create,

let's just clean up all of this. Now it's giving me an AI thing. Emphasize the idea of recording on capturing existing con content rather than generating new, okay? It's given me a bit of a [00:29:00] bit of a thing. Gary Chu is not on the first page here. Neither is Nathan Barry

talking about. I documenting events. Okay. The concept is the same, but it hasn't given me the same reference points. I think Google's lost the plot here. It has not picked up that article by Nathan Berry, and so chat GPT is showing us that it has the ability to script the internet much faster. Now granted, I use deep research and I use deep research pretty much.

Well, I wouldn't say all the time, but when there's something that I know where it's like, wait, I'm looking for new information. I'm looking for information that's relevant today. I'm looking for context for what's going on in the world now as opposed to just gimme general information. I'm using deep research.

So I dunno. I think this is probably one of those hacks as I was going through this process and as I was [00:30:00] journaling this morning and think about, okay, capture not create. I was thinking about course creators because I've tried this process where it's like, okay, I have a course in my mind. I wanna put out my education.

I wanna create something and I wanna make it great, or whatever. You try to build a course, but building a course and recording a vid, recording videos, editing them, getting, you know, an editor involved, piecing it together. Buying a platform or subscribing to a platform like Thinkific or Kajabi or something to host the thing, maybe school.

And then it's like, okay, upload. And now there's this thing that is stuck in time, right? It's like publishing a book. It's stuck in time. It's, it's one of those things where you have to create it in a way that is evergreen. Now, evergreen courses are great. There are principles like CAP or Create that.

Going to work. You can build a course around capturing create, because this is gonna be a thing that human beings are gonna wanna do all the time, [00:31:00] is to figure out how to solve this problem of content creation. And how do you do it naturally? How do you do it authentically? So yes, there are universal principles in the world like.

Music and philosophy and human behavior and all, all a bunch of things that are relevant and that are evergreen. Those are all being published already. The most of the content is free or in books. There is this shift, or there is this idea that I have this concept where because of AI and because of the way human beings.

Have this need to constantly have new stuff. Video formats change, editing formats change, quality of video changes, quality of audio changes, quality of technology helps us do things better. There's this constant need to go after the shiny new thing. I just went to get a book from my bookshelf because, as an example, the format of a book has been the [00:32:00] same.

For ages. Essentially the inside of a book today, 2025, looks very similar to what an inside of a book looked like maybe a hundred years ago. This format works. This format is evergreen. This format is timeless, but courses are not because of the things that I just mentioned. And so here's what I think works, and I'm gonna be testing this.

I'm bringing this up now in this video because this is something I want to test is. Instead of doing a course that is static, do it as a cohort three times a year or four times a year thing. Maybe things change in your industry. Often you need to do it more often. Good example of this, Emma Porterfield or Marie Folio, they have these schools or they have these platforms where they're, they do this community, they do this cohort, and it's a model that works because.

What they're doing is they're refreshing their information all the time. Advertising standards change tools that we are using change the way that we do things [00:33:00] online, changes the way that we are interacting as human beings changes the way that I've just recorded this video and shown you my chat GPT screen and how I'm thinking about what HGBT is doing for the future and how it's interacting with us.

Has changed from what I thought about last week. Just content in general has got to have that newness to it because that's what people are looking for and because, well, yes, you can do evergreen content that's searchable, but the context of today will change. And so from a course creation point of view, I would recommend that instead of.

Buying courses that are static. Pay a little bit extra and join cohort based courses where you get access to the community. You can ask questions. You can obviously interact and you can learn the same stuff that you would from a normal course that you're just watching videos, but maybe that course is old and you didn't have the context, or maybe that course creator did not [00:34:00] update it for a while.

This is the trouble, right? If you're a course grader, this is something that I recommend that you do. If you're a student, then this is something that I think you should think about too, is when you're buying a course. Ask when it was last updated figure out is this relevant? Are the topics contextual to who you are now to what's going on right now in the world?

Because yes, you could buy a book and it could be timely or, or timeless. Right? Keep us on influence by, it's already 11 years old and, uh, it's still highly relevant because it's, it's built on the foundation. Of like you wanna be a ki for influencer, influencer or an influencer, but he's guaranteed he's going to write different books with the same content, with different titles because he knows that he needs to change the book cover.

He knows that he needs to publish something new so that he himself is relevant. Yes, he'll update it with certain things. This is the thing that I think we should do, is. Work on things that are, that [00:35:00] have a balance of being timely. The principles are maybe evergreen, but the context is new is today. And even when you're making YouTube videos, I heard this thing this week, a book by Brendan Kane, where he talks about hook point, where he spoke, he speaks about in your content creation journey, publish a video, publish something, then look at the data.

Then figure out what your next post should be rather than batch creating. And this is the thing that I've gotten into as well. This is why I asked chat GPT, like how do I go with this caption, not create concept. Because when I batch create, the first video is normally okay. The second video is okay. The third video is usually the better one, because I'm now in flow and I've created, I'm sitting and batching for a while, but I usually find that the videos in the middle or the videos at the end are probably the best ones.

But then I'm like burnt out. Then I'm like, I don't wanna sit in front of the camera for [00:36:00] like weeks and then I lose my consistency because I'm like, oh, I went through that whole process of batch creating and now I don't feel like this is important anymore, versus what I'm trying now is, okay. This is pretty long episode, and to my editor, thank you for taking the time to edit this video and piece it together with the screen recording, but it's like, well.

How do I do this in a way that is more natural to me and that I don't have to worry about my hair, that I don't have to worry about the quality too much, but it's like, let me share what's on my mind right now. I can figure out what the title is gonna be and I can maybe take a screenshot from the video or I can maybe, you know, work on the thumbnail, uh, in that way.

So I am going. To share my thoughts. If it's daily. Okay, cool. If it's not daily, okay, cool. It's like, okay, these are gonna become podcast episodes. I'm publishing these videos too. Very simple thumbnail design. I have obviously photos that I have inside of Canva that I [00:37:00] can just slap on a, on some text, but it's fine because the thing that I care about is not so much views or, I mean, sure I want growth, but at the same time it's like I'm just sharing my thoughts.

I'm just capturing what I am noticing, and if I'm wrong about that, I probably will be wrong about 70% of the time, but I'm okay with that because I'm human, right? We are expecting AI to get it right. This is the thing. We are expecting AI to get it right, but as human beings, we're forgiven because we have that ability to forgive.

We're not gonna forgive if my fan that's spinning around. And keeping me cool in this hot environment in Bangkok. If this fan stops working, I'm not gonna forgive it. I'm not gonna say, oh shame. It's okay. You've been working for 12 hours every day for the last year and a half. I'm just gonna throw it out and buy a new one.

I'm [00:38:00] not gonna forgive it. I'm going to replace it Anyway. I appreciate you for listening. If you want to check out what I actually do for a living, uh, there are links on this page. Uh, you can go check it out if you want to consult for my business or if you want an order for your landing page or your ad account, then definitely go check out my, uh, links in my bio or links on my website to be able to book a call with me.

If you just wanted to have a chat, like DM me on Instagram or wherever you can find me, probably Instagram is the best place and we'll figure out from there. So I appreciate you for listening. If you've gotten all the way to the end, comment below and let me know what you think of the stuff that I've spoken about.

If you've observed things separately or differently, then definitely, um, correct me, I'm open to, to your interpretation as well. Good. Thank you so much. Take it easy. Have a good time. Have a good weekend, joy.

Bye.

‍