#27 GPT 5.1, Webflow AppGen, MCP's SUCK, Quitting AI Coding

Is that my sound? We are live. Wait, am I using the right mic? I think so. You think so? I think so. One, two, three. Can you hear me? Is it like from this mic? Yeah. Yeah. Technology is working. Yeah. Technolog. We are live on Instagram. Welcome to the Technology Show. Technology Command. Uh, I need to set the um Hello. Hello. artwork for the live stream on the YouTube. Cool. My light is way too white. Is your white balance off? Yeah. No, I changed the light bulb. Yeah. I don't have like expensive soft I have a soft box, but it's not like an expensive light or anything with high CRI, you know, the CRI index.

Um, no. So, all lights uh have a CRI index. And if I'm not wrong, this is like my this is like from old time when I was still busy with filming and etc. Uh a high CRI index essentially like 100 is basically the the color quality of the sun like the render that you get out of it. The quality that the sun has is 100. And the basic lights that you have, like the basic LEDs and stuff, they are probably like under 70 or even worse. So that's why when you film yourself or take a photo with a camera, you your skin looks like pale or purple or whatever, like green. It's because the light quality is so bad.

So what you want, you want to buy a light with a high quality. Is it CRI? I might even like not be sure about this. CRI does sound a bit weird because I know about the color temperature, you know. Yeah, that's different. That's different. So, the color temperature is completely different. CRI, what is Yeah, it is CRI. So, CRI uh rating uh from 0 to 100 with higher values indicating better color rendering capabilities. Wow. So 70 to 80 is typically sufficient. Google says no. You need something better than 90. Like even probably better than 92 or so. Anyway, so I have a bad CRI color and the the better ones are very expensive. Like the the sky panels from Ahri, they cost like a light cost like 6K.

Just one single light. Yeah, but you could get I mean just for YouTube, you could get a pretty good light for like a fraction of that cost. Like I've got two I've got two lights. I've got like a quite a big I mean it's not one of those boxes by any stretch, but it's like a I don't know. It's probably about what do you say nine in 7 in something like that. And then this one's like a tiny one. This is just a little little but I got two. Yeah, they're good. Yeah, but but these are small and and the bigger the better. Speaking of light source size and also camera sensor size. But yeah, because the bigger source is softer like on your skin.

Yeah. Yeah. Yeah. I want to get one of those light bars, I think. And I want to put it behind me there just for a bit of don't know. This is what I did. This is what I did. So, uh, you see also the wait this way. You see how the light I can turn it on and off with my phone. And also I can change the color temperature. This is actually this is Yeah. What do you that light the light inside of Okay. Yeah. Yeah. You see how the color temperature changes? Yeah. I I like it a bit warmer. But the one I have in front of me is colder. So if I make my background too warm, the camera tries to balance it.

So I get too blue. Yeah, you're very blue right now. Yeah, I know. I know. I'm basically avatar. Uh, and then you see those like small lights there. They cost me 15 bucks off of Amazon. Yeah. And I can look, I can change their like intensity and I can do a bunch of like weird with it. Like I can change it's a light strip. Anyway, yeah, I think I did change it and now I don't know how to fix it. I have a show in my background. Yeah. Uh, static. I I needed static. I don't I don't want a gradient or whatever. Yeah, I I literally did. Okay. Uh, let me quit some stuff so I'm not all jittery.

Fixed it. Okay. So, Jimmy, yeah. Should we start? Should we start? I was going to ask, do you have any interesting or fun or, you know, news to share with the world today? Uh, wait a second. Wait a sec. I was trying to read the comments. I'm sorry. So what you were saying? Anything? Any news to share with the world today? Any news? No, not really. Been working on a huge project. It's live, but I I prefer to share it like on the from the official channels uh in a few days when we have a few more renders out. And yeah, been doing some really cool stuff. Oh, exciting. I'm I'm surprised this is not an official channel.

I thought this would be an official channel. This would be the official channel straight from the horse's mouth, you know. Uh I have things to show by the way just to say like but not regarding like the agency stuff. Mhm. Um yeah for the agency we are like building the the whole our whole new website from scratch with the our branding. We are approaching it from like a 3D perspective. So we are like using Blender to do our branding essentially. Uh very cool. Yeah. I'm trying to move my website because I want to change the the language around what we do because I I I'm I'm I love building apps. I freaking I really prefer it I think to building websites.

I used to be like no I love building my I mean you know with my brain it will change every which way every other month sort of thing. But right now I'm really loving building apps and I'm really enjoying and AI has got a part in that but like I'm I'm trying to build this like we'll get into it later but just trying to build a really robust robust robust coding system so that you know I can work efficiently on my own. So sneak peek like I signed up to code rabbit to do code reviews for me, you know, really leaning into the fact that I don't have someone to like check my blind spots and things like that.

So yeah, I'm I'm sort of investing in a lot of tooling so that I can work more efficiently as a solo dev using AI. So, um, so yeah, long story short is that I I need my my company website to kind of reflect the sort of focus on building apps and I want to take it off WordPress because while WordPress is free, hosting on Hostinger is still something like 120 20 quid a year, something like that. That's a lot. Okay, maybe not. Maybe not as much as that. Maybe that's an exaggeration, but either way, I can I can host it. I can get it on Versel and host it for free. Yeah. Or netify. Yeah.

Yeah. Well, I'm using Nex.js and I'm using payload CMS to uh to run the website. So, yeah. Um it's a lot easier for me to update if I'm just AI coding it. I can send claw code for the web off to do updates and things like this. Like just makes sense as I'm a solo person really leaning into that that um Yeah. I makes it easier for me. Uh we also got asked in the comments what we will be talking about. Uh we'll announce that soon. We have like a bunch of things. Web flow app gen AI GPT 5.1. We'll talk about that. But what you started with changing direction and stuff. Uh I'm interested in that conversation at least for a bit.

Uh we are rebuilding our agency website and one of the verticals that we are adding we are adding verticals meaning that which type of clients we are trying trying to serve for example middle or big client bigsiz companies uh startups I don't know like companies like this we have a bunch of like verticals what we are adding which we didn't have like last year is AI driven companies because it seems like there are a bunch of them and there is a bunch of like money going into these AI companies. This is what I want to get into. I want to do this but at the same time I'm like hang on hang on. Yeah, exactly.

Kind of like okay when well if it pops and if there is no money in it we can just turn it off or you know we can still serve the ones that survive because there will be some surviving ones. Uh so a project we've been working on um for the past month or so and we've been like tirelessly working on is for an AI company. So this company I don't want to reveal too much yet just yet. I will have a full video about it is taking like AI and doing something very specific with it. And when you do something like this you need like very specific UX and UI. And we've been doing the most of like the UIUX designs and also like the assets like all the 3D and like visualizing things and making things beautiful and explaining things.

We've been doing all of these things and that got us thinking like why not serve this industry uh even more since we now have like experience doing these kind of things. we built like some um systems around it and things like that. So, we will add that vertical, but at the same time, what if it's it's gone? I mean, you know, it's no big it's no big deal for you. It's just something you offer. Um I mean, I could talk about this quite a lot, but it was something that really really hammered home, right? Because in amongst all of this AI, I mean it's whilst it it may or may not whether it pops or not, I think it's completely going to reshape everything, right?

And certainly as a developer, it's going to reshape my job as a developer or at least there, you know, I'm just spitballing now, but I think it's definitely opened up a new segment to app development, right? Which is this heavy prototyping phase, which doesn't need handwritten hardcore code, but is, you know, functional and, you know, useful, right? I think there's going to be a whole new vertical. I'm kind of thinking about this on the spot which is which is quite useful. But saying that um in amongst all this like oh what's going to happen and this and that and the other uh I was I was like you know at lost right for lack of a better word and I was like well what can I do?

what resource do I have that are available to me and uh how how can I move forward in this unknown new landscape and on one aspect I was like let's retrain it's focus on the AI stuff learn all that on the other stuff it's it's exactly this it's like shaping the company and and knowing which vertical we go for and blah blah blah so I was like who who have I got available to me and I landed on a specific vertical because I have lots of friends and contacts in that vertical and I just sent them all a text and I sent them all emails and whatever saying, "Would you mind having a conversation with me?"

Not really knowing what was going to result in the conversation. But through having really deep, luckily I know these people well, right? But through having really deep conversations about their industry, what started to take shape was a really solid idea of the some of the challenges that they were facing. And that was really it was an eye opener because I've always known this. I mean Ran's always talked about it. The future these are all YouTube channels that that we've have various affiliations with and stuff. They've always talked about like niching the riches are in the niches and all this kind of stuff. And I sort of finally started to see it. And so that's that's kind of where I've been focusing my attention on.

I've I've actually got some pretty good movement in that space. I don't know whether I want to transfer that to my lap uh my laptop my my website. But saying all that, one thing I've really seen is um I watched a video on uh it was I watched this travel YouTuber called Simon something or other and he goes to this uh billionaire uh hotel and it's like eight and a half grand a night for this hotel and he comes away and it's you know for our for Plus, very lackluster experience. Like, it's fine. I would say the staff are a little bit too casual for eight and a half grand a night. I would say that the services were okay.

You know, very nice. Don't get me wrong, but was it worth eight grand? And you see that in the comment section. You see a lot of people being like, "That is not worth eight and a half grand. That's not worth eight and a half grand." And it sort of taught me that if not everything is made for you, like you you have very very very different priorities to another person. What that 8 and a half grand is buying you is not a fancy service which is what we look for when it comes to holidays and things like that. If we're going to be paying a lot of money, we expect a lot. What that eight and a half grand is uh you're paying for is privacy.

You've got Kim Kardashian going to that hotel. You've got billionaires finding um what's it called? Uh when when you're alone in sacred or serenity, okay, you're buying that privacy. That's what they're paying for. That price tag, you know, keeps the riff raff like us out of that hotel. So, their priorities are very, very different. tying that all into what we're chatting about is that like not everything needs to be for everyone. And if you if you don't see or understand the value on the one hand, you know, as someone potentially offering that service, there's no reason to take any offense to that, it's like, well, it's it's not for you. But to be able to do that, you need to really really understand the the people you're trying to serve and their their pain points and all that lot.

And this has really solidified in the last week. So, it was funny that you mentioned it cuz it's re that was that was a I literally watched the video like 3 days ago and I was like I'm kind of really understanding that it's fine for so many people to not want what you offer, you know? Yeah. Because it's not for them. They just what's the anal? There's something about like you're just you just look you're looking for something completely different or you're offering something. You're not offering what? Yeah, there's some there's something in that. I hope that makes sense. It does. And yeah, uh we'll try it and we'll see. You need to brighten yourself up because you're in the shadows or there's something shadowing o over your Yeah, it's better.

You're a bit blue, but you know, you're you're not in the shadow. Oh god. The problem is uh Okay, let me fix it. You know how Oh, it's hot. Let me know in the comments if any of that made sense and if you have any comments about it because yeah, might be something to consider because we can't be everything for everyone and it's okay that people don't want us cuz different people are looking for different things. I'm literally unscrewing my light. Oh my god. And it's too hot. Oh my god. What? What? Is it not LED? Yes. Jesus. That's a big mother. Yeah, it this is really good, but it the the white balance sucks.

And now I'm too yellow, I guess. I don't know. Well, it's okay that you're I think you'll find that you're yellow, but it's just that hot spot on your for That's all right. That's fine. Probably better than me now. Shame. Right. Should we do the intro? I've had my rant. I've had my rant today. Actually, we've got tons of rants. So that's not my last rant. I'm allowed more. Yeah, we we do have like a bunch of stuff. Always flying, always always thinking, you know, always. Yeah, we we're just having this because we had this, you know, before even going live. Hello, Chris. Hello, Saiya. I see you in the comments. All right. Um, so this week we have uh Okay, you you're ready?

I'm not ready. I'm always ready. I'm always ready. Okay. Born ready. This week, Web Flow launched their AppGen, a tool that allows you to create any app in Webflow without writing any line of code. Just by prompting to AI, you create any apps. This is revolutionary. We'll show you a few uh examples in a second. Uh, OpenAI released chat GPT 5.1 and Anthropic is saying that MCP sucks. That's the creator of MCP literally. So, that is going to be interesting. Uh, this and much more in this week's uh AI. Uh, I'm in in this week's command AI. I completely messed it up. Um, I'm Karza and I'm Sam and this is a week in AI design and dev.

Yeah, I completely miss. All right, so we have Tim. Hello, Tim. Uh, cool that you are here. We are going to How does it sound? How does this week's episode sound to you, Tim? Let us know. Well, he we are going to show his project. Oh, okay. Yeah. uh we're going to show a project he's been working on uh using the web flow appgen. Also, if you're double screening, if anything pops up because what I've found is that we'll get off and there'll be something like happened whilst we're online. So, if something comes up, let us know and we can talk about it. Yeah. Live. So, but first we have the website of the week which I want to grab the link again and we'll get into that first.

All right, I'll share my screen and we'll go in a second. So, we need to remove your screen, sir, first. All right. So, for the site of the week, we have a super cool 3D website, and it's very um minimal. That's what I love about it. And here it is. So, this is We Create Feelings by Mantis. Is it uh a studio? I think it's a studio from two people. We'll check out uh their about page. So, they have this really nice intro and you see this is like 3D and you see kind of like the veins and they're not the veins but it's a very cool effect. And did you see how it loaded?

We'll see more of that. We create feelings. So, this is about storytelling. And as I scroll, you see how it's all 3D and it I just I I really love it. Um, great design doesn't just guide behavior. Then you go and scroll. It speaks directly to the nervous system. Ah, so it's not veins, it's your nervous system. There you go. Yeah, it's really cool. And it's kind of like starting with a touch like the, you know, that's really nice. And you see it's kind of like procedurally being made and drawn even though it's 3D. That's really cool about it. Uh we activate nerves. This is this is a nice tagline for your agency website. Like you have a client like we'll talk to the nerves of your clients.

So yeah. And here they show up uh like these really really nice visuals. I I really love them. The minimalism in here is really beautiful. Yeah, it is. And do you see like the details, the typography here and the color green? It's just there and it's popping but it's not too much. Uh same here. And then here they have New York City and then they have the weather. We are adding this to my agency website. We are going to add this. I'm not sure if we will add it in the he in the nav or in the footer, but we had this idea to add it. A bunch of agencies do this. And this is just so cool.

Just like nice little details. And as we scroll, look how the next thing pops. And yeah, I absolutely love it. We stimulate memory just like very short taglines. And then I mean when you have Adidas and Google and Microsoft and Meta in your portfolio, you don't need to yeah to say too much. Uh except they have also awards, not just that. Uh and yeah, this is this is super cool. And this one, look how nice this is coming together like the the effect that they are creating here. And I was just thinking about how I would recreate something like this because it's always really difficult when you've got like height, right? Animating height in Yeah.

Well, you can animate how he height to auto, right? What I what I don't know how I would approach and this is what is interesting me here is you see how these are dented, right? And if I click on development, it's going to come up and it's going to shift the rest as well because like as they come out, well, this one doesn't. Well, if I click, no, this one. No, they have to be Yeah, like this. You see how they come up and then as they come up, they shift the rest as well. So it's kind of like the elements are being triggered to be full width of what they can be once they pass the threshold.

So maybe is this like a Gap observer? I'm not sure. H they are using Gap and Versal and obviously 3JS for the background. We'll talk about that. So, I don't know. This is This got me thinking I should recreate this as a component in Web Flow. It's nice. I love just the gray. Gray and green is such a classic. Well, anything like orange and green, gray and green, like when you've got this, it really makes an impact when you see that green. I'd like to see more of it in some ways, but like um they've you know when you see the green as um yeah, it draws your attention, you know. It does. And this footer is also pretty nice.

What's all that you up? Yeah. And the the hover interactions and this hand. This is pretty cool. By the way, am I lagging sometimes? Like if I'm lagging, I I should mention this. My M1 Pro, I can hear its fan. Oh gosh. This doesn't happen a lot. So I project Are you plugged in? Well, yeah. You expect too much of your little Mac. You expect too much of it after that 50 50 watt charger. Yeah. So let's actually check this for a second. So uh I do system information and then here we should see the power. I should see how much and this is kind of related to the website as well. I want to see if this website is causing more power to be drawn because I mean obviously we're live streaming, right?

Um but where do I see scroll down to the bottom? So it should be there. I think it's there where before it starts to indent. The last one. 52. Yeah. So, the V it's 52. And I'm pretty sure this thing cannot like properly get 52 watt to my laptop. And that's probably why I'm lagging. So, I have to close up the project after we are done with it. But, we're not done yet. A few other things. They have an about us. And the about us kind of changes the the mood a bit. It becomes all colorful. And I'm honestly not a big fan of uh what are these uh creatures called? Praying mantis. What? Praying mantis.

Praying mantis. I think I'm not sure what mant Well, I like Yeah. I like I like them, but I don't like the art style and I don't like the I don't know. I don't know. I feel like I think these are real photos, you know. It could be. You know what? It's I think this is called like an orchid mantis or something. Okay. Cuz it looks like a flower. Isn't it sad that we are asking this question? I know. It's crazy, isn't it? It is quite sad. Like because of AI, we can't appreciate the photography and art anymore. For real. We can't We can't Yeah. Leonol says uh that he likes the color or she.

I'm not sure whether you're a boy or girl, but uh you do. Well, I I do like the color. I I'm not I don't have a problem with the color, especially this one. This is really nice. But I feel like going from that gray, grainy and green to this, I like the idea of contrast. And I do also see this logo and kind of like the the claws or like what are these called? Like the hands. Uh I see the resemblance, but I don't know. I feel like what they had on the homepage is so special and beautiful. Yeah. So, yeah, the mood change is is a bit is a bit jarring. Yeah, but you know, it's just it's probably just me.

And when I say uh when I click on contact, this thing pops up and it says hello. Yeah, it's kind of cute. Um, but yeah, back to the website. We see how it is. And this 3D, I'm not sure. Do you think this is this is probably a 3D model made in Blender, right? And then pulled in into 3JS. Yeah, I'm not too sure. For people who are interested, I'm just learning this, but if you go to 3J 3JS.org, so this is 3.j 3js.org. Um and here you can go to editor and in the editor you can actually upload uh a 3D file. What is the the format called? Um there are like a bunch of formats that are supported and then you can play around with it right here and then obviously you can take it from there to build something like what we are showing.

Uh but there is a lot more uh that goes into it. Things like properly optimizing and this site even though I thought it is optimized it's causing my fans to run. So yeah, but still pretty cool. Very cool. Yeah. So, uh, anything to add here? No, just, uh, it's a very cool site. What's the website? The website, we'll link it below, but it's mantis.works. And I kind of want to check one more thing on this website like going to okay so there it's a one pager essentially like well there have the contact but here are the services these are all links to the other uh sections of the website. Yeah, presumably it's just a normal website with a canvas layer behind this fixed position behind it probably, but you have to like observe the scroll so you can you know animated and then uh if we look at the code.

Yeah. So these are the sections but I'm not sure where the canvas layer is. Maybe here. No, this is not. It's probably easier to find it here. Not sure where it is, but yeah, it is definitely C a canvas. And yeah, and it's pretty cool. It's pretty cool. Uh they won awards and I think they they deserve it. Oh, they won awards for it. When when was this when would they release this? I think it was last week I saw it on awards and I was like damn I like the vibe. I I like the uniqueness. And you know something special about this website like if you see it again you will remember it or like if you see any like resembles lens of this somewhere you will you will get it because it's it's unique.

It's very unique and that's what I like about it. So presumably they are a kind of design UX studio then. Yeah. Yeah. They they showed in their services uh what they do. They do a bunch of things like uh 3D stuff. You see experimental digital 3D branding identity. They do also development. I'm suck I'm a sucker for like a concept like because really the website is pretty straightforward like it's again it's just a canvas normal website scroll down one pager that kind of moves this person but this concept of nervous system touching and feelings and whatever I'm always a sucker for like how it translates how that concept that idea translates to the rest of the website it's really really Cool.

This is my favorite part, too. Yeah. Um, how would you do it? How would you how would you come up with a concept and then how do you try to translate or like expand in this case you expand it to to do more? Yeah, honestly I think it's I think it's like a brainstorm exercise creating like a mind map trying to get to the core of what it is that you do like and start and there's probably a lot of word play that comes along with that like engage touch feel emotion and trying to and then you know through kind of uh what do you call it just sort of a process of of elimination just coming coming with some values or some birds that that you feel quite positive about and then trying to establish a relationship between them.

Like maybe emotion and touch uh like touch points and engagement are all, you know, possible with the with the nervous system or something like that. You know, a lot of it is just like, you know, pulling your finger out and just like making up something. But that's creativity. That's exactly what creativity is, right? is just coming up with those correlations um that sort of tell that story. And then it's a design exercise. Then we like this idea of a nervous system. Maybe there could be a ner like some um veins around the site or like lines. Everything's made up with lines, you know, and they've not overdone it, you know. It's just they've they've they've they've hit on that concept and and haven't haven't just gone really super heavy with it.

They've just Yeah. Just is what it is, you know. So they they reduced it to one thing and I like this uh so open AI which we talk about a lot on the show. Uh they they they had this they have this like very generic logo um and they came up with this branding and they made a video about it with dots. So it's essentially dots that come together and create shapes and create things. And uh they even have this in Atlas in their browser where there are dots that are going around. So I I'm fascinated by this like and again this is like the something that you do in branding and when I mentioned that I'm doing a rebranding for our own website doing it in 3D in Blender we're trying to come up with ways that we want to look at things.

So one idea that we are working on and it's not final is blooming. So that we help businesses to bloom. So again worth play. So you go from blooming to flower. Um but we want to make it 3D. And uh if you make a flower and make it 3D it's it's a very feminine thing. Uh and that is nice but we also want to be kind of like h go in a techy and more sharp way at it which is not feminine if you do it sharp. So we're trying to balance things out by imagining how can we make a flower that is 3D that is also not AI looking um because we are going to make it in Blender if we go in that direction and then we are trying to basically come up with these like concepts and then these concepts kind of bloom into bigger things and then one thing that we are working on is like having the flower uh to be kind of like feminine and character and soft and beautiful.

Uh but then kind of like in the shapes that we are using in the website to use like sharp corners and a really red like extreme red that I have in my branding. So a bunch of things like this. Uh and then obviously we can animate it. We can make it bloom and things like this. That's that's how I'm looking at it. And for clients, uh, we try to find like what is that kind of like a memorable part of their brand or their experience and kind of take that and bring it down to something very fundamental, very simple. It has to be simple because when it's simple, then you can scale it back up to do things with it.

Like the dots in OpenAI, if it was something complex, you couldn't do much with it. But because it's a dot, you can make things with the dot. And here with the nervous system, they kind of connected it to feeling and making memories and kind of making you making the design pop without saying we make the design pop. We were saying like we will touch the soul of your Well, they haven't said this, but that that could be a line that they could say. Yeah. Yeah. Yeah. Yeah. Yeah. It's fun. again if you can and the only difficult aspect of it all is like you know there's always going to be this difficult balance between we want to we want to resonate with our target audience we want to make sure that we say the right things and we use the right colors that really you know resonate resonate is the only word I can think of right there but at the same time you still want to you know stand out you still want to put your personality across so there's this middle point I think that's very very difficult to reach where you you're considerate of your client your target market but also expressive enough where you're a you're you're not just a do you know what I mean like you it's very very difficult to find that and if you can do that that'd be great.

Yeah. Cool. Well that that went into branding conversation when I and I like that. Cool. Uh should we move on to the next one? Web flow app gen. Yeah. All right. Well, uh let's see. I haven't got this, unfortunately. I haven't got it. You haven't got this? Oh, they said I have. I got an email just before this show, actually, and I haven't checked. Maybe I can check now. But yeah, was very confused. I was even recording a video and uh that little button didn't show up. Okay, now now at least I've got a notification. I didn't get I got a notification but it wasn't in the notification list. So, okay, it's not there. So, yeah, share your screen.

Yeah, sure. In a second. So, the most Web Flow No, not the most Web Flow. The most powerful Uh, okay. I'm going to redo this. The most powerful Web Flow feature is out and that is AppGen. A feature that allows you to prompt your way into creating any app you want. Things that were simply not possible before in Webflow uh or needed, you know, de a developer to do the whole development for you. Now any designer can just prompt and create a fullon full-fledged app. Uh and here you see the the environment. So I've been trying to create a dashboard for my clients so they can log in and add tasks and see their project and the you know the status of their project and add a bunch of things like this.

This is very early. I just did a few prompts and I haven't been really working on this. Um, it's been like 20 minutes, but uh, even with that time, I got it kind of like, yeah, like a login page. And after the login page, I know this is quite ugly, but we'll get into this. You have this I like it. There's something I like about. Okay, cool. Well, the red is I'll show you where the red is coming from. But the idea is that here I can add new tasks and I can, you know, add an image. Um, but a bunch of things are not working, which is a given. I'll show you a better example of what I have.

But yeah, this is the UI. And when you create something, let let's go step by step. So yeah, you go to your web flow project. This is my web flow project. This is my whether it's a full web. So this is obviously a a web flow in production web flow project but it can be an empty project. Exactly. It can you can start from scratch. The reason I didn't start from scratch and I did it on my agency website is because I have a bunch of CMS and component and variables. So if I go to my styles, I have some colors. This is old. That's why it's not really sophisticated. But I have some stuff.

I have some components and I have some CMS items. Why does this matter? This matters because now you have this appgen and when you go to the appg gen you can now I can create another app by the way you can create multiple ones. This is where you start. You start from scratch you can add a prompt and you can add your collection. So for example I have career h and job location and job types uh collection. And I can add those. And I have also components. I can add my global footer like nav components, things like this. And I can prompt the AI to use these components throughout the app. So the app, unfortunately, this is what I don't like about is a very separate experience, right?

It's completely separated. Well, I'll show you where it's connected, but it's for the most part separated. you prompt it and then it starts creating it in a sandbox and then you can deploy it. Um, and by the way, all of this is still in beta. Uh, and the thing is that it takes those variables, those colors and you know design system that you essentially have and reflects that inside of the app. But you still have the option to go into your files. this is like your files and uh presumably do a bunch of stuff. So, you tell me about this because I'm not very experienced, but it seems like I can I can add code.

I wish I could see the clo uh the file tree more. Are there any style files in there? Uh good question. So, we have what is webflow CSS? But it's something that they can't change. Yeah, because they're defined inside of thing inside of Webflow I guess. So we have inside your own styles. So where would the style live? Oh, any any Yeah, there we go. Um because I'm just so just to this is all this is all Webflow cloud. So previously you can you can get developer to publish a a web a react website to web flow cloud and bind it to your web flow website. This is a way of not not having to do it inside of web flow cloud but to actually do it directly within web flow.

Um and this will use Astro to do it instead of Nex.js. If not familiar with Astro you can write code, right? You can edit this code. Yeah. Yeah. Here I can't re uh I can't edit this code because this is a footer component and look how interesting. So if we go to cover.com for example and then if we go to my footer you see here here I have like this footer and here these are this is like essentially my footer component that is being pulled in. So we have these site components that are all here. We have the styles and we have a bunch of other things. I apparently we we should be able to also deploy this um well we we can only deploy it to web flow cloud but we can also push it to GitHub and I don't know what we can do there.

Like can we take it from GitHub and do something with it? This is what I presumably. Yeah. So what I'm understanding here and I'll be interested to know what code gets written on components that access your CMS. But that aside, what you're seeing here is a React component of your footer. The way to achieve this in Webflow Cloud is via Dev Link. And the way to get your styles inside of your Web Flow Cloud project is via Dev Link. So it's like again it's sort of installed it's created an Astro project and installed Dev Link and pulled in those components through through that Dev link system. Um I'm hoping that makes sense that this is a basically just a web flow cloud project using Dev Link but you've you've created it inside of um inside of Webflow.

The fact that you can push it to git is really really cool because then yeah you could push this anywhere you could because it's using dev link which doesn't need I'm pres I'm assuming now doesn't need to be inside of web flow to be able to use it. Everything should be good. Yeah. Any chance we can see any component that's using your CMS collections? Um well let me show you another project and then we see if how that one is waiting for that. I'm waiting for that ability inside of Web Flow Cloud to be able to read from a CMS and use I don't have it I don't have it in this project unfortunately but uh all of this was done with essentially one prompt.

This was one prompt and the subsequent promp prompts were to fix bugs and things like this. Uh but a good friend of mine Tim Tim Schulz he's making this for his client. And check it out. Like if I reload the page, you see it has even like some animations. Again, like all of this with just prompting, but he's been doing this. He's like he spent like six hours uh prompting this. So still everything is uh local. So it's on my local storage. But look at this. This looks really really good. So he said like he was like very specific with the style and prompting and being like do it like the Apple style and uh apparent apparently for the big part this is like functional.

So like for his client they can look at this. This is so good. This is so freaking detailed. It's just so feature. Like it's crazy to think like a designer, not a developer, can do this in Webflow and this is like on his client's domain and it's not like a subdomain or anything. It's just like on the main domain or on the Web Flow domain. You can you can do it uh the however you want. And look at this. This is like really good. It looks really nice. and it has a bunch of like information and presumably a lot of these are coming from the CMS. But yeah, he's been prompting his way into this and I and I find this to be very very impressive.

Like it has even settings and these are like set to soon. This is very professional. Yeah, it's really good. So yeah, just a reminder this is like was not really possible at all. It wasn't even in Webflow's wheelhouse to to support app you know they ignored um what was the JavaScript library that that was made by or bought by whizzed you know that was the only way to build like an app experience using web flow. It's not quite the same because obviously designers can literally design the pages inside of Web Flow that later gets transformed into a web experience. This what you'd have to do is you'd get your designer to build all of the components and and literally create them as components and then then you can prompt them into your app.

But um yeah, it's not much different um honestly like designing it literally on the canvas or like in the component. Well, it is different but you won't have that fine tune. you won't have that fine tune and I wish for that and I really really would love to see this appg gen to not be a separate thing or like the preview to have you know the CSS capabilities that I that web flow has so I can click on stuff and change them and it just changes them in the code but presumably if you change the component inside of web flow that will just update in appj yeah I don't know because I don't have this component on my website.

So, this is completely made up, right? It's kind of like Yeah. But I I need to do more and, you know, testing to see that. But from what Tim has been creating, I'm looking I've been very optimistic about things that can be done here. Yeah. Yeah. I mean, you know, I don't want to look at this. I'm I'm I'm very happy it's here, but it's not going to be too different from Lovable Vzero, all of that. Yeah, but it's just The thing is, yes, it's not going Well, here's the thing with Lovable. Lovable has the experience to create all of this, but they don't have the, you know, the platform infrastructure what Web Flow has.

You know, Web Flow has the CMS, the editor, the designer, the capabil lovable cloud. Yeah. Well, yeah, but it's nothing like, you know, what Web Flow has with all of what you need for a marketing website. Uh, now you in Web Flow, you have all of the Web Flow platform stuff. You have animations, you have Gap, right? You have that that amazing visual builder. On top of that, you have Lovable essentially. Yeah. With Appgen. Yeah. Yeah. And just just a reminder as well for people, it's like this is this is an Astro project. It's I I'll have to have a look at it, but it needs to be like the assumption here is that it's not a serverside rendered experience.

It's a it's a JavaScript powered experience. I don't know. I'll have to double check. I mean, in theory, if you upload a NexJS project to web flow cloud, that's a serverside rendered experience. But this, you know, this is really where the use case for this isn't so that you build a website just from vibe coding. You can totally do that if you want to sign up to Web Flow, start a new website, and just vibe code the hell out of the entire thing. Obviously, you can do that. Yeah. But I don't think you'd be getting it can be quite frustrating to vibe code a website. You don't get the fine control that you expect. Um, Web Flow's created a good compromise though by being able to import components that have been built in Web Flow with Devink.

Great compromise there. But, um, yeah, I think there if you just wanted to vibe code an app and that was it, then there I think there are better systems out there. But if you want to vibe code an app and incorporate it within a web flow website, this is this is great. This is amazing. Yeah. I really hope that they will open it up more. So all of that had no not just that like the code so you can do a bunch more more with it. I'm not sure where the limitations are to be honest but knowing web flow they haven't been the most you know pro open source or something like this even though they have like the code export and stuff to be uh to be fair.

Uh but yeah, we we never have been able to, you know, have a back end like access to a back end per se. Anyway, but in theory, did you see how cool this is? Like I added this. Yeah, it's amazing. I'm just so surprised. This is so good. Yeah, sorry. You were saying um so this should use Web Flow Cloud, which we covered last week. This should use that. Sorry, webflow storage whatever data storage this should use that so you can access the database and things like that um I don't suppose you have access to this project no this this is um for now all local storage but he's going to add it add a bunch more uh he he has a full plan for this but still this is amazing I mean presumably you could you can vibe code a link with um sort Superbase or neon.

I don't know how environment variables are handled through this but they are available in web flow cloud. So assuming that that would work but yeah you could you could use web flow's built-in uh databases or you could just say build me tie this to uh to a um what do you call it if you're going to add a feature. So can you uh go into the prompt thing down the bottom right and just type at or can you reference files at all? No. slash slash or anything. Yeah. Yeah. So, you'll really you you want to be able to say update this file to do this, that, and the other. Do you know what I mean?

Well, yeah, but I did. So, I you see like from here there is no UI to create a new Yeah. So, they really just want you to stay in the the window. Yes. Yes. May But maybe they will add it because this is very like Yeah. So, I said I didn't know what file to add. I said just add a cover ts and add it into a folder called kverza and it did. So I prompt my way into adding a file. No description provided. How are they how are they charging for this then? What's the pricing? Uh from what I understand it's nothing for now. But uh I don't know because like you can't do unlimited AI, I guess.

So you have I I think you can add your own um your own AI API to it. Yeah, sorry. Your own API like for like open AI. Oh no, like for your AI like the AI uh costs are I'm not sure because I don't see it from here. Uh but if we go to site setting I don't know if I will be showing some yeah so here yeah so if you go into that web so this is your application and then go into main so these are your environments so here these are your deployment see it says environment variables don't don't click on it because in case there's something I don't know what would be in there but next to that you've got bindings this is your database.

So should I show this on the stream? Yes. Uh bindings. This is your databases. So whether it's So this is where you would manage your databases, right? So I can add a storage and it says skew like database key value and object storage. Presumably if you prompt it, it would just create these unless um you told it not to use that unless you told it to use superbase or whatever. Um it would create storage in here. Um but yeah, but this is yeah this is again hopefully it would ask you for your superbase API key this that and the other. Um and so yeah this is pretty cool. You could create a like an actual MVP or even a potential product with this, right?

Yeah. Very good. Yeah. say good. So just a few now that we are talking about web flow uh very short updates. Now I got a notification that I can do multi- uh design design with multiple people on the canvas at the same time. So that is out. We've been talking about it but it wasn't out for everybody. It was a slow roll out. Now I have it, but my team is not around right now to show it. That's unfortunate, but because I wanted to really show it, but yeah, whatever. Uh, and comments. Apparently, I can do comments and see them and and see them even now. Yeah, that's really cool. So, you see I'm in the design mode and I still see the comment.

So this the comments are previously you had to toggle constantly and that was not really nice. Well, you press C, I guess, but now you see it all the time. So you can just be working and see the comment. You see a comment. Yeah. Yeah. Um and look how cool this is. It shows what breakpoint viewport size. It has a bunch of metadata that comes with the comment. So was that left from within uh web flow? Was that left in the staging environment? No, I just did it now. I did it like this. But I can share the project in a comment only mode and share it anywhere and then you can Yeah. By the way, uh Tim, uh the creator of the other project that I showed you, he's saying in the comments, you can add Superbase as a database, um user management and storage.

I have tested it and added uh at the end. Yeah. Yeah. Yeah. Again, that's the thing with code, isn't it? you can do whatever like the it's impossible for a system to say well it's not impossible actually that's stupid but like it if you can write the code we'll get into this later on but the code will run what you write so if you tell it to go read from a database from neon object storage from S3 authentication for superbase it has to do it there's no way that really web flow can stop you doing that you know which is again probably why they've avoided it for so long is because It's it's the what do you call it Pandora's box almost you know once you allow them to do anything they can't control what you know you can't control what systems they use by the way why is there such a like huge lag when I type something here I see things on my screen with a lag normally terminals in a browser tend to be quite laggy because you're you're sending a SSH connection to an to a terminal.

Um I I don't think this sort of thing has the idea of like websockets or things like that. I think you know you you you open up uh Digital Ocean and try and open up a terminal session in Dig Ocean and there is literally like a lag where JavaScript you know. Yeah. Okay. Yeah, that makes sense. That makes sense. And yeah, uh just so you see like to deploy you you see how you can change the the path. This is really cool. It's you know it's on my domain slash whatever path that I want and then I just press deploy and it's building the app and deploying it. Very cool. And I can access my variables from here as well.

So it seems like maybe maybe Appgen will be its own UI and Webflow will port more from like the design to here or it's a two-way binding, you know, you can create cuz Yeah, you you That's what I mean. Yeah. Yeah. Yeah. Pretty cool. Nice one. Should we move on? Yeah. Try and steam forward. So, OpenAI just dropped their most advanced model yet, and it seems to answer um or bring back what's been lost for so long, which is empathy. People have been at a loss without GBT40. Will we have glazing again? Uh probably. We'll see, actually. So, GBT 5 5.1's been released. There's two models that have been released. Uh there's instant and there's thinking.

And I think they're trying to change the terminology a little bit. We've seen thinking quite a lot as a replacement for reasoning. Um and instant is kind of the more it's a renaming of the chatbt model. It used to be called catchbt model, but yeah 5.1 instant 5.1 thinking. This will be the one that you mostly use. I don't know why they say on most used model when it's just been released, but heyi ho, you know, maybe they're referring to the major version as opposed to the minor. Um, now warmer, more intelligent and better at following your instructions. I have a feeling this is this is really where it's changed. I wonder if there's much actually happening that's different in the thinking, but they just need to make sure they're both aligned and whatever.

But anyway, most advanced reasoning model now easier to understand. So it's definitely a personality thing. Someone called it, you got IQ, I think it's EQ, like emotional, you know, intelligence. Yeah. Um, faster on simple tasks, more persistent on complex ones. So really, I think you need to know what you're doing to to to uh to be able to select this one. I don't have a chat GBT subscription so I can't really show you. But what we can do is take a look here at the differences between GBT 5 and GBT 51 instant. You you can definitely see this is very cold to the point. You know, some people might prefer this, but off the bat you've got I've got you wrong.

That's totally normal. And you know I I remember this. Just shut up with this. I I don't want AI to be like confirming. I'm watching a video on this. Yeah, the majority of people are going to want this. I quite like it. It makes it at least, you know, I work on my own a lot. It makes it at least a little bit more thing. It's definitely going to be divisive, but I think they announced it or they previewed it or something. I think Sam Alman said in a tweet that they're tweaking the kind of customizability of your model or making that sort of front and center. Like I think you need to tell your model because this is going to be the majority like want this approach.

You need to tell it to be um you know reduce this you know remove emojis, be a bit more straight to the point, don't glaze, all that kind of lot because you're in the minority there, right? Yeah. Um but either way, you know, you can see it's not really tremendously longer. It's actually shorter in fact. Have you seen the video? Sorry. Sorry. Ch made me delusional. No, I've got it saved. I've got it saved. Um, does it I mean in the middle of it, well, almost done with it because it's one hour, but it's pretty fun. The whole point is like CHP and all AI is a language model that try is trying to please you in a way or another.

So even if they don't glaze you, it's a machine that predicts the words based on what you're inputting. And these like subtle changes that these companies are coming out with are essentially like changing the knobs to be like more warm or cold or friendly or like alienated or whatever, right? So from what I understand this this is just you know a a bit change in that sense and it's not a revolutionary you know a product or like much smarter um in anyway sure and yeah that video I really highly suggest it watching it um chat GPT made me delusional by Eddie Borbach it's quite funny and you get a feeling for okay what is cha.

It kind of give you like a very strange but like interesting perspective on what these machines are kind of by going with the flow of what the AI is telling him. You get a sense for okay is there are they actually smart that that's where you will what you will ask yourself. Yeah. Uh we'll get into some of the kind of certainly from me like the question I I do as much as I sound enthusiastic about AI, I still do challenge myself. I still do challenge my own understanding. So we'll get into that a bit later on in the show, but um yeah, I do want to watch that. Um but at the end of the day, this is just the new version.

I mean, it's not necessarily right or wrong, but this is kind of the new version of social media. Social media just wants you to engage, right? It just wants you to um have have that dopamine hit, want to return to the app. It's gamification in in AI form in making it, you know, more appealing to to use it, you know, to I I wouldn't say that this is a new concept. They're never it's never a new concept. You can always draw parallels with the past. All I think is that that, you know, it's fine to for a for an AI to be a bit more like this. I don't know. Everyone's going to have their own opinions.

saying that the thinking one um it's still this hybrid model where it adapts you know it's got this auto rooting which you know historically hasn't been so great uh people really struggle with this idea but you know you're putting your trust into it not overthinking so uh when it just needs to answer you a simple question like how many hours on strawberry and all that kind of crap um but you know it's hard to really demonstrate that for in this kind of stream. Um whether it's you know what I mean whether it's taking ages to think or not. The other aspect that we haven't that that we've um that's not in here that got I think got released today or yesterday there's two new codeex models as well.

There's codeex I think it's codeex mini and codeex thinking or something like that. So yeah, another another I I like to use I use kilo code and I use GBT 5 mini for like planning and stuff like that. Uh I I'm be interested to use codeex now 5.1 codeex for planning because I've heard that that's quite good for yeah just planning out your code tasks and whatever and then executing you tend to use a lighter lighter model. So, we've hit 100 views. H 99. If you're on YouTube, if you're on YouTube, hit that like button. I'd love to everyone to hit that like button right now, wherever you are. Or if you're not on YouTube, scan this QR code and come join us on YouTube where we're having lovely conversations with Tim and Leonel.

Leo, by the way, how Yeah. How can people scan it if they're on the You can take a screenshot on iPhone. You can take a screenshot. Oh. And with that screenshot, it will pick up. Oh, cool. I didn't know, but I learned something. And next week, we've been promised by Streamyard, which is the tool that we use, that we will be able to for you Instagrammers out there, we'll have vertical. And for you, you know, I don't know. Should we do widescreen or vertical for Twitter? That's the question. Um, that's a good question. No, I would probably do also vertical for Twitter. It's still It's still pretty mobile, isn't it? Yeah. But yeah, I man I I went bonkers to do this with uh OBS because you can do this with OBS.

You can do like multilayout streaming, but it's so freaking difficult like Yeah. Yeah. Apparently, I don't know what it I guess we set we set up different layouts cuz like if we go to this layout, you know, it's got to put us on the top and then whatever and then on this layout it's got to stack us. Yeah. Whatever. Anyway, we're getting we're getting too nerdy for our delightful viewers. But anyway, just to wrap up on the 5.1 thing, always welcoming new models. It's not, as you say, it's not going to be the groundbreaking model that we um hope and dream. Although, I will say I think a lot of people will appreciate that warmer, friendlier approach uh while they used AI because some people aren't out there just trying to get work done and get on with their day.

A lot of those people are using AI to interpret things and to understand their feelings and whatever. Um, so yeah, we'll see. We'll see. I'm not a big user of chat GBT, but I will be moving on to 5.1 codeex in kilo code just uh just to see how the planning the yeah the planning uh agent works. Cool. Anything else? No, not really here. Cool. Wow. Go on. No, I just said uh I'll try to uh I I saw it in my cursor, by the way, but I'm out of my credits. I'm out of money, even though I run through my like 30 extra euro credits as well. So, yeah. Uh I I I'm I'm constantly like running the auto model right now.

Well, look out for a video I'm about to release probably next week now on Trey because I've really I really really enjoyed using Trey today. It's half the price of cursor. I know you've paid for the year so you're screwed. Yeah, but Trey, I've really enjoyed using Trey. Um and from Can I move my project over? Yeah, it's just Yeah. Yeah. Just open it with Trey. Mate, you need to get out of this web flow locked in. Yeah, I know. I know. walled garden. I thought you're in the code world now. Yeah, I do. So, um, so last week like we I didn't know like with uh Versel that you can't do a team on the free tier.

So we I moved to Versil like I I not moved like I deployed my project on Versal and I asked my team to join as well and deploy and then I noticed that I can't do a team on the free tier and I wasn't like willing to pay just yet because we're still exploring stuff. So we now moved stuff to Nestlei to do to have a team and so we are we're kind of doing this and and then we are like after we built our project we kind of like rebuild some part of it in web flow again which is not really the best flow right uh but yeah that that's how we are trying to build these components for for our tool but you can use def dev link though de it's not what we want to do in this case like in this case I can show it in a second just a teaser uh before okay so here uh I don't have it so I'll just run it and just build it and Okay.

Run dev. Um, did I do it? And npm run. I typed too fast. Okay. So, we have it. And yeah, so this is how I'm using cursor right now. Yeah. Yeah. New browser mode. Yeah. I love it. I love it. Especially right now, uh, I'm sharing my Mac screen. I put what you see with typically uh with I make the chat a bit bigger. I put all of this in my big 32 inch screen and I just fully vibe and I do a lot of like chat um ask mode as as well. So I literally go into the the code and I don't do the heavy stuff yet, but I do like uh some variable changes or something like this.

Um I'll do it I try to do it myself but I this this project is way beyond my like scope in order like because you see it had I have like the editor of the runtime I have like a build process and this was given to me as a prompt by my senior developer. So he gave me a long prompt saying like build a project like this because before that I had it just as a one single JS file with an HTML file and just merging them. And once we build this I ship it to uh Versal or Netlfy and then everything that I have here is built in a way that works with the web flow project.

So the idea is that you have these mini tools and you can like do animations um the way you want. Yeah. Yeah. Yeah. Well, you know, you're in the land of code now. You can you can move this onto tray easily and uh it's you're not locked in, you know. Yeah. Um and it is a VS Code fork at the end of the day, but um it is. Yeah. Yeah. Yeah. So, it's it's going to work with all of the nuances of VS Code. Um, yeah. Anyway, should we should we get on with the comments? Yeah, let's go talk about the comments. So, this is the comments section where we take a look at what you have to say.

Uh, let us know what you think about anything you hear down in the comments and we might just read it out. So, I'm a little bit active on in uh LinkedIn. Maybe a little bit too active, some might say. Basically, you love LinkedIn. I don't necessarily love LinkedIn. I think it's definitely a place where to get work from for as a developer, as a you know, whatever. But, um, you know, so just just as a note, I should probably share I have shared this if you want to follow me on LinkedIn or whatever. I have shared this but like everything that we say tonight will get transcribed and broken up into a individual post that I then schedule out on LinkedIn.

Completely AIdriven workflow. It's incredible. So last week we spoke about uh PewDiePie and uh it's got a lot of attention on LinkedIn. Are you gonna It's a bit awkward, isn't it? Oh yeah. Got it to stage. Yeah. I've been seeing a lot of conversation around PewDiePie overall in the dev community as well. So, this is interesting and I want some input from the from you guys right now because this is something I'm trying to adjust to. So, what I loved about PewDiePie's journey is not only, you know, he's he's obviously a very successful YouTuber and doing silly videos and gaming and stuff like that. We said all this in the last week's video, but I love this journey that he's gone on actually becoming pretty much a developer.

And what I really love is how AI has empowered him to one, it's inspired him, but also just the age that we're in. It's so easy. We thought the internet was great, right? having the internet in our pockets with our phone. But now with AI, we can get very bespoke learning and we can learn new things and and and much much much quicker. And in fact, we probably could do it more. Like I'm learning a lot of songs on guitar at the moment. I've yet to pick up AI to even approach how I might do that. Maybe there's something there. I don't know. I've always told you about, you know, learning whatever. And you know, we got 22,000 uh impressions, uh you know, nearly 100 um likes, repost, comments.

It's it's picked up a lot of steam. But what I'm finding is that let's take this comment as an example from our old friend Paul here. It helps that he's a multi-millionaire with time to do whatever he wants. Now, I interpret that as and there's a lot of comments around this that there's uh you know, that's all all right. He all it started was a dream and a personal fortune because bearing in mind the the the stack that uh PewDiePie built was like 20 grand, wasn't it? It's something crazy. Yeah. Really really expensive thing. But I don't really mention that really in the post. I just sort of talk about his journey and whatever. But there's a lot of like focus on this idea of like needing cash and I'm trying there there are a few people in here saying, you know, that they're constantly learning uh what they've really enjoyed about it.

Someone's here is telling is saying that they've built their own kind of LLM and doing this that and the other. A lot of people were inspired. Um it's just a few and we've spoke about this briefly before the the show. Few just seem to focus on the the financial aspect and I want to know if that if you've seen the video whether that's what you got from that like the the money aspect. I I want the focus to be that now is such a great time to become a developer. it's so easy um and you know the resources are out there to become an incredible developer and that shouldn't it should not stop us but so if you got well actually I'll say because it was your article wasn't it that you you sort of brought to the the show what was your like I watched the video and I was uh inspired I was inspired to do more like I see PewDiePie doing this and I'm like amazing this guy I've been watching ing his videos for a decade now.

Legit like I don't know maybe 12 years I've been since I've been like watching his videos and yes sometimes less sometimes more. Um and I saw like he's becoming more interested in engineering and this is amazing. It's it's it's not just you know somebody liking engineering it's somebody getting deep into something right. Yeah, in this case this is engineering and computer science and it could be I don't know it could be uh something else. It could be anything essentially he had a time where he got into drawing by the way. Yeah. Yeah, I saw I think I saw that but yeah I Yeah, you saw he mentioned it in his video as well but he I think he had like 100 days or like weeks of drawing.

So he had this is the thing about PewDiePie by the way the one of the reasons he's so big he has so many followers is he has been changing like getting into things and like actually giving it a try. Not just you know liking it for a day or two but actually trying it and sharing his experience whether that's gaming, drawing or at this time uh development and computer science and engineering. And it's very uh nice to see it. And I don't think you need a crazy setup to do this. Probably most of people in the comments they have like the minimum enough to do to start with something. And even if you don't have that, even if you have just like a very basic laptop, you can fire up a cursor, right?

And you can start learning. Yes, you can't run your own LLM, but it doesn't matter. you you you can do something else. Front end, back back end, any type of learning, you can do it. Um, so yeah. Yeah. And I'm glad that's your sentiment, too, because again, like I want to just empower you guys who are watching the show to be, you know, not looking for the negativity and not looking for excuses, which I feel like are some of these comments are looking for excuses to like not learn or, you know, it wasn't the fact that he built I mean, I was slightly inspired that he built a ser, but I wasn't about to be like, oh, he can only do that because he spent 20 grand on five, you know, I don't know what the name of the graphics card are.

No, I was like, what can I do on my uh fouryear-old, 5-year-old MacBook M1? Um, and you know, I was inspired in that way. But it's interesting that people by the way like most people who have the money that PewDiePie has don't do the things that he is doing right now. So, yeah. So, that that that's not really the point. The point is he is learning something. This he doesn't have to do this by the way. He like he he could easily like hire people build him an app or you know build him whatever that he wants but he's not doing it for the sake of that. He's doing it for the sake of learning and having fun with it.

And that should be your takeaway. Yeah. You pick something that you absolutely suck at. That's also a part of it. You absolutely suck at it and it is going to be painful at the start, but you pick it and you stick with it and you go, you learn it and while you advance, you will have fun with it because you will have fun with it once you get good, once you like get something and you can make things, you will love it. You won't as long as like at at the start, but just know it. That's a part of the process. trust the process and pick it up and yeah, go go learn something. And we'll get into this again later.

We are queuing up such a big chat right at the end of this show by the way because I've said this so many times. We're going to talk about it later on, but this is queuing like I want to, you know, let's not let our instant gratification, the instant gratification world we live in, uh, deter us from persistence and, you know, persistence, right? It's going to be hard. It's going to be tough. Um, and we aren't going to learn things. We aren't going to get things right straight away. But slowly but surely, you start to see, oh crap, like I'm actually learn. And you'll be surprised. Reflect on what you've learned. Reflect on where you start.

Remember what you were when you start uh when you started out this journey and just reflect on that and be be encouraged to get better and learn more because now it's it's easier now more than ever just to kind of like start something, not get immediate results, and be like, "Okay, I'm going to move on to the next one." just this stuff takes time and especially with development but design there's so much more there's always more to learn just just be encouraged that it's easier to learn and don't the I guess don't rest on the laurels of like um just because it's easy to do something with AI say coding dig into the code understand what it's doing and that's what we'll talk about later on but yeah I think we've we've spoken about this long enough but I just want our community to feel uh inspired by what is going on rather than feel let down or discouraged because that's what a lot of people do and they bring a lot of negativity to to some of these comments.

It's like they're misinterpreting the point. Um and if they're going to if they're going to make excuses for not doing for not being a better version of themselves and let them be it. Um yeah, similarly back on LinkedIn uh share this tab only a little one. We won't dwell on this one too much. um is the affinity uh post. So this is uh it's yeah so just to touch on my post from the the link from the thing Adobe should be very very worried um canvas obviously offering the free version. I don't suppose you've used the free version have you just yet? Not yet. Um I touched on it a little bit. It's uh it's cool.

Um and one thing we didn't speak about actually that a lot of people raised in the comments um obviously a lot lot more positivity um in in the comments here is well what happens when things are free right when the product is free you are the product which led me to a little bit of digging and Affinity and again we don't want to be too pessimistic but Affinity released a video specifically calling out a few things. Sorry, I'm just accidentally going to repost that. No, we don't want to do that. Um, we've been listening to your comments and there is one big question that's This is the Could you hear that by the way? Yeah.

Yeah, let's because I haven't seen it either. Okay, it's on one and a half speed. Uh, hopefully it'll be all right. That's fine. Yeah, that's fine. Inity is now free. How the hell is that even possible? It is a fair question because for decades professional creative software came with a hefty price tag. Subscriptions, lockins, pay walls, that became normal. So when we made Affinity, some people assumed there had to be a catch. Let me be absolutely clear. There is no catch. We don't sell your data. We don't train AI teachers on your files. We don't monetize your creativity behind the scenes. Your work stays yours always. So why make it free then? Because we believe every designer deserves access to tools that respect their craft.

Because the industry needs a shift from gatekeeping to generosity. And because we've built a sustainable business model at Canva that allows us to support this kind of access without compromising on product quality or user trust. This isn't a marketing stunt. It is a long-term bet on creative freedom. Affinity has always stood for fairness, choice, independence, and program power. Affinity and Canva both share the mission to empower the world to design. And making Affinity Free is our way of pushing that further and getting creativity into more hands in more places than ever before. Everyone with the Canva premium plan will be able to access Canva AI and other select Canva features within Affinity. But the Affinity app that you all know and love with all of its professional grade features and updates will be free forever.

And all of this makes sense for Camber as well. When more pros use Affinity, more teams end up using Canva. Pro designers lead the visual workflow. If they build an Affinity, they're more likely to scale in Canber. And when they do, marketing, sales, and the rest of the business will follow. That is how it grows. We know trust is earned. So ask questions, kick the tires, hold us to a high standard because that is exactly what we would do. Affinity for creatives by creatives. There you go. So just to just to recap, no data, no training or anything like that. Um want to empower designers to to you know do what they do best and um they are assuring us it's going to be free forever.

I like that they because I'm you know I'm still relatively pessimistic on some things and you know whilst I can see this is obviously you're a business right? You've got to earn money. Um but they do actually touch on the the the fact that is um uh that that business like marketing and sales will follow if their designers are using a tool. Do you know what I mean? Like the money will follow and it all stems from the creatives. However, the one thing that I think that we want to remember is that so and we we have already spoken about this actually so many businesses use Canva. It's crazy actually how much um how many businesses use it.

Some massive massive companies already use Canva. So if they can incorporate or encourage their teams to you know utilize something they're already paying for there is that. And also finally just you are paying for those AI features on the back end which is becoming really crucial and they're not gimmicky AI features in these sorts of software although it does have image image generation. We're talking about content fill and cleaning up of images and things like that. We're not talking about just horrendous, you know, nano banana changing stuff, although people will still use them and more power to them. Um, this is tasteful use of AI, I think, you know. So, yeah. Rounding off. That's really good.

That's really good. The transparency is amazing. Yeah. And they're a British company as well. I was quite pleasant to pleasantly surprised by to learn they were British company. Cool. I didn't know. Yeah. So, we got some comments. uh change to WordPress from web flow, never look back. See, it's just one big cycle, isn't it? You know, yeah. Um yeah, I mean I am moving off of web WordPress just because whilst WordPress is free, it's the hosting that costs loads of money. Like there are much more efficient like ways to run websites now. But like yeah, web flow WordPress can be a great solution for a lot of a lot of people with very very flexible and a lot of freedom.

So yeah, more power to you. But yeah, um Ken, let us know what made you move off Web Flow. Um and then uh I don't know how to say that. Probably Cynthia Cynthia Banks, Scythe Banks, something. Um we just doctor Oh, about to talk about MCPs. Give them a different function. They have other benefits other than context or use some feature. Do you know what? I don't know what you're talking about there, but we are going to talk about MCPS. Yeah. In a second. Yeah. So, I saw this video from Theo that basically says Anthropic admits that MCPs suck. And they do suck. They eat up context. They they're unreliable at best and they make your AI dumber because again that additional context is causing a I think they call it um needle in a haststack problem or something like that.

people have realized it and that it's a standard that needs reinvention and who better to make that realization or to admit that realization than the very people who invented it and thought of it. So yeah, uh Theo made this video just kind of basically going over this article which raises some really interesting points. Again, it just admits to the the tool calling aspect. Constantly having these tool calls in context whether you're um whether you're using them or not. I it's a headache to be honest. While I'm coding, I'm turning MCPs on and off and then I forget to turn them on and off and you know it's just it's something you just don't want to have to manage.

Just a quick side note, I did learn because I'm starting to use Code Rabbit. I'm trying to build up a bit more robust um coding environment that picks up on my blind spots, allows me to be a more uh impactful independent programmer. Uh I install Code Rabbit, which is an AI code reviewer. Um and they get you to use the CLI through clawed code. I didn't realize that you can use CLI through core code. So as an another example, Taskmaster has a CLI. This is internal. This is boring in terminal like you know uh you type what would you do in terminal? Do you use any CLI that we could you know not really dumb down reference to but yeah instead of what do you mean?

Yeah I want to simplify it. I want to simplify it. Yeah. Yeah. But um I'm not there yet. I don't use Yeah. Uh I mean you're you're in your terminal instead of having a guey having a having a UI. Yeah. In my terminal I just do get stuff withinal. you've got git inst. So you've got git you've got git installed on your computer and you you trigger that CLI with the term git. If you start that terminal command with git that's using now the CLI and they're the arguments that follow that. I hope that makes sense. I didn't know that. So okay that makes sense because I do get status right yeah to get to understand if I have pushed or not I forget and then I do get add dot to add everything essentially.

And this is a great example because you can ask your um AI to push if you just say push this to git, it will do git commit and and write a nice commit message and get push. So it I always knew it could use a CLI because it's using the git CLI. Uh you can use it with code rabbit. Similarly, I use taskmaster which eats up so much context. Taskmaster has a CLI which I've got installed. So now I'm just going to remove that MC. have removed that MCP and I'm going to use the CLI. That's just my own personal thing. That's nothing to do with this article, but it's a way of me now just really um you know realizing that MCPs just suck.

Okay, they they just I don't I I prefer to preserve context and not have to manage stuff uh and have my AI not be stupid. So, here's here is uh Anthropic going through the downsides. They're not going to completely tell you it's rubbish. They still keep their kind of one foot in the door with MCP. They still want you to use it. But what it does go on to say, and what I've noticed recently, actually, so it's it's funny how it all comes about because, by the way, this was released, I think, on the 4th of November. Yeah, 4th of November. What I've noticed that AI is doing now is writing JavaScript files to do sort of push stuff to to the database or to do certain things.

So it will write its own code and AI has always been great at writing its own code. So what what claw code is suggesting here is set and I've yet to understand how to actually do this. Do you ask Claude to create this MCP server? Um, but ultimately you you tell it that there is the MCPS are here as files that it can use. So you can remove the MCP and this just uses the APIs um to to access Google Drive instead of the Google Drive MCP. there's a bunch of files there which files don't eat up context unless it's actually going to be using them. So then the AI will write a piece of code that imports um actually this is the wrong one here.

Uh it's this one. It imports the Google Drive files and then runs this bit of code and then this is this is the only context that it's using, you know, because it just runs this code and then it's gone. It's not going to be wasting up. And what they found was 150,000. This is wild, by the way. 150,000 tokens to 2,000 tokens, a time cost saving of 98%. What are you doing anthropic? Why have you why are you still using this MCB this standard that's that you can if if you can rejig it just a little bit you know maybe this is a skill maybe this is something like 98.7% is crazy crazy so I want to start looking into how I can incorporate what they're doing here I'll I'll read the article a little bit more deeply but and how I can inevitably create a you know a file-based MCP service or whatever if indeed Google Drive doesn't have a a CLI that I can just tap into because a CLI will just run.

But ultimately, you're asking your AI just to run a bit of code. Um, so yeah, they're admitting MCP is is not so great and there are other ways to use it. But in the summary here somewhere, they again they kind of keep their foot in the door. But ultimately, um, yeah, MCPs are they on their way out? I think people struggle with the concept of them. I would love it if they work if they were more deterministic if they would always run. I've I've resorted to saying use the taskmaster MCP to add this as a task or use the linear MCP. You know, I've been very It used to be that you could just lightly mention, you know, add this task or um what Figma like uh take the design or something like that and it will just know to use the MCP.

But it's just too hit or miss. So you might as well just use use the Figma MCP to you know get that node or whatever it is. So anyway, all that to be said uh checking on the comments here. Anything to add? No, not really. We have a comment. Yeah, exactly. Yeah. So I thought MTV just talk to the same language through uh our apps or that was the internet. I think what you're saying here is that um the it is kind of it's just kind of like a layer on top of the APIs, right? You're talking about APIs here, I think. And uh yeah, MCPs are just a way of for for the AI to interact with APIs inevitably, but they always they always stay in the context of every single request.

So each request just gets larger and larger. it kind of compounds and your price the the cost of that thing compounds the intelligence of that request compounds you know or compounds in a negative way. Um well another comment here I wonder about using MCPs as attributes for AI as needed. Well, this is the thing. You you want to use them as needed, right? You don't want them just to hang around forever in the context when you don't need them. So, um so yeah. Anything to add from your end? Not really. I'm I'm not really You don't use MCPS, huh? No, they are I'm not sure if they are here to stay. Yeah. Well, you know, I don't know.

Yeah, AI are an interesting point at the moment. Um, I think the we'll get into it later on. Again, I'm queuing you guys up for a banging conversation, but yeah, the indeterministic nature is just whatever it they are. It is quite nice that you can again just ask something to do something in natural language. That's great. It really is. But there needs to be a way to do it without um you know having to uh constantly I don't know it just they just don't need to be there basically. So yeah. Yeah. Um let's move. We won't do clawed robots unfortunately. We'll move past robots. Let's let's talk about what I've been chatting about this whole episode, this whole show.

Yeah. Um what I've been queuing you guys up for. So this guy CJ, uh last month he kind of lost his love for coding because of AI. He was he was finding he was basically becoming a manager which famously people just don't like to do um become managers. He missed actually coding so because he was just managing his AIS and stuff like that. So he made a vow to like not code for 30 days and he actually made this review of kind of what what he's learned and I followed this journey just because again like I said earlier like I'm someone who does want to keep challenging their own beliefs and see the other side and start to you know question whether I'm you know rightly informed on everything.

I I don't want to be just kind of blind to all of the the shortcomings of AI. Yeah. And I kind of wanted to point out a few of the things that he learned, right? Um, funny enough, even though he had only done 30 days, he still found that he craved using AI for quite a few things, right? Um which which is interesting because you think that because he really missed actually manually coding and he he did a few what do you call it like he did a few um projects and uh which really tested his you know ability inability to to use AI or lack thereof of his project. So, um, one thing that I really like is that he talks about this cycle of AI slop being now put onto the internet.

And I I really don't want to see this coming because a friend of mine sent me a video that was AI and she thought it was not AI and I was like, "Sorry, that cute dog is is AI." Um, and I I it made me think like if people are constantly being bombarded with AI stuff, they're just not going to use that platform. Instagram being this example. People are not going to use Instagram because it's like every other post is faking AI, you know, or they need to be able to filter that stuff out. The internet, I hope, is not going to be the same thing. Like if every everywhere we go, every website we turn to, there's just AI crap on there.

there's no new information or anything like that, people are going to stop posting to the internet or stop using the internet. I really hope that's not going to be the case. Or someone needs to come out with some big filter that says no to AI stuff. Like even these AI browsers, I'd hope that we could turn a thing off to say just don't show me or don't suggest AI created content that there's some sort of interesting detection. I would love that because Yeah. Yeah. What he turned to is a new website that I found now called Devdocs. He's turned to using documentation above something like Stack Overflow, which traditionally we would use just because the the question has already been answered somewhere on Stack Overflow, right?

Stack Overflow driven development. He resorted to using dev docs or documentation just in general. This is just a nice way to have it. So this will you can download and install actually any documentation to any sort of language vit uh view uh what we got it you know webpack all this sort of stuff you can actually select those for offline re like referencing so if you are getting into development like being able to and god damn it documentation can be ugly you know it's very very difficult to sort of pick apart But saying that, this combined with something like, oh, you can't see it. I've got my I've got um I've got a comet open here with my AI assistant.

I actually wanted to create an MCP wrapper for this website. I don't know how whether it's possible to do that, but um yeah, some way to to be able to reference documentation or bring that in because I have MCPs for documentation. See long story short is he found that he found a lot of um answers to actually understand what's actually happening by reading the documentation. So that was a a nice little insight. Um he also found that the human aspect to defining the infrastructure of applications was was something he used to just throw to AI. I'm still at that level. It's kind of weird because I kind of understand how applications are put together. Uh, and I think that's what makes me a good developer in the age of AI is that I can trust that the AI can do the the the uh, what do you call it?

The manual stuff, but I'm there happily overseeing it and making sure that everything's roughly done in the right way. But to define and design like the architecture to your app, you know, your your thing, right? You had a developer put that in a prompt, but they were there to think about how you would architect this project. That sort of stuff and they didn't do it, by the way. No, I know you didn't do it, but you Yeah. Oh, what? You didn't you didn't use that prompt. No, no, I used the prompt and the prompt had uh like exact steps and like an exact structure and they did like 80% of the the things as it was told to and the rest just blew it.

Yeah. It Yeah. didn't do it correctly. Wow. Yeah, that's the thing. Like you can Yeah, you I would have Yeah, exactly. Right. I probably would have just given you a boiler plate and that's enough for it to like there's plenty of boiler plates out there. Yeah. The amount of like prompts that I go through to to get things done and I change stuff. Uh sometimes I just add features and then remove them and I because I don't do it in a let's say sustainable way. Um it it gets really busy very fast cuz I'm building like MVPs essentially, not Yeah. very clean. Yeah. Yeah. Um so I mean that was that was that it was it's interesting to see his journey.

There is there is still you know he's a hardcore coder. There is still a time and a place for coding. But then I saw another video and I won't we're not going to go through all of this video but this is a cold fusion video. We've mentioned a few times on the on the show. This was a video just released this week and it talks about the kind of you know are we really ready for AI? It used to be called vibe coding but they must have updated the um title or something. But you know just reading about this it's kind of like a a semi-spon to you know lovable sponsored this video and I don't care what who you are.

If someone gives you money you are going to be slightly biased to that person. He does say they don't have any input blah blah blah. You are going to be slightly biased. I'm sorry but it's just that's just the way the brain works. Anyway, obviously you're seeing a lot of lovable stuff here. this still lovable stuff. This was created in lovable uh this was created in lovable a lot of lovable stuff. So anyway, you know, again, just kind of Oh, he even mentioned CJ's video too, actually. So, it kind of made Yes, he did. He did. I I watched this video as well. So, the whole video is sponsored by Lovable. Um, but the video is basically showing how unreliable AI is.

This is my uh take from the video that yes, you can build a prototype. Yes, you can do a bunch of stuff that you couldn't do before. Uh but now if you want that polish and you want the things to work, you know, properly very often AI can't get you that far. Um and this is kind of like a starting point and yeah, it's AI is not going anywhere and we're not going to use it less. we will certainly use it a lot more. But I think what needs to be settled down and you know we need to kind of like get there and understand that it's non-deterministic. It's it's not really as reliable as these companies are trying to to say and it still requires an actual expert who knows what the hell is going on in order to create something proper.

Mhm. Yeah. And on top of obviously saying go watch these videos for yourself and you know come to your own conclusion. It's it's left me in a really interesting point because CJ mentioned something which is and you've just said it as well is is the undeterministic nature of AI and coding is the most deterministic thing you could probably ever do. It will not do anything that you don't write. any bug that's in the code, you can't blame the code. You can only blame or sorry, you can't blame the execution. You can only blame the code that you yourself have written. It is so deter. So you've got this massive dichotomy now of like well I'm trying to create something that has to be deterministic that has to do exactly as I tell it to using a tool that you can write the same prompt and it's it's different every time.

And this is something that's really it's really affected me the last you know week or so or half a week and that I love doing this. I'm trying to again like you say it's here to stay but I'm trying to figure out where this will find where this will rest where's the final resting place for AI driven development or have I exhausted all of my options when it comes to AIdriven development and that's why I picked up Code Rabbit that's why I tried Trey I'm trying to um and one thing that I want to share is I spent I think we need It goes back to what I was saying about not being lazy and just um you know just because AI is writing it doesn't mean you shouldn't delve into that the code.

I've really I really don't want to get sucked into moving fast and and building things um and using all the extra time that I've now saved to just build more features. Like you could it's very easy to get in that rut of just like well now that's taken me three minutes let's move on to the next feature. I really trying to utilize the extra time to make sure my um my system is set up as best as it can. So again, setting up rules, setting up MCPs, although I'm moving off of those now, having a code having code rabbit run and and all this kind of stuff. I'm I'm feeling a bit fatigued and feeling a bit frustrated with the with the indeterministic nature and I'm trying to make it as deterministic as possible within reason.

But I think we're going to really struggle if there's coding is going to really struggle if they don't sort out. I would argue the indeterministic nature of AI is worse than hallucinations. I'm just going to say it. Like I honestly think that hallucinations can kind of humans hallucinate hallucinate right they have problem but but at least they what they write or whatever you know what they say goes do you know what I mean I don't know how they're going to sort that out to be honest right but this is kind of where I'm landing on with the AI stuff I'm not going to stop I'm just going to try and get better but it's just like further understanding where AI is going to stop, you know.

Um, not a coder or innovator. Would you guys We got a question here. Would you guys recommend I knock out MVPs then hire dev to clean it up? I don't vibe code. Um, yeah. I mean, like at least at the very least to have fun. Do you know what I mean? Just understand what what the what the limitations are of uh of vibe coding and stuff, you know? Unless you never want to be a developer, unless you have no interest in doing all that, then just hire someone. But if you if you are interested, follow the curiosity and then buy code something and see where you get to. But there will be things if you have no idea if you if you don't code, there will be absolutely things in that codebase that cannot go live because you're at risk of leaking people's data, getting hacked, do you know what I mean?

All that kind of stuff. So yeah, that's all I'd say. Any thoughts? It was a long rant. I told you it was going to be a long rant. Yeah. No, I I I I think my stance are here clear. Just uh I don't think like coding is going away and uh there is an infinite amount of nearly infinite amount of code to be written and software to be created and problems to be solved. So I would say go out there, use AI to learn. Uh this is what I'm doing. Learning is learning is the best thing you can do with AI. I think the best thing you can do with AI and code obviously, but learning is great.

Um do we have time for this last one? Yeah, let's do this last one. Uh I think it's not a necessarily long one. It's cool. Apple intro. Are you going to take us through it or shall I put it? Uh, no. You can put it. You have the I don't think I even Did you? So, have you guys seen the new Apple TV intro? Sorry. Um, I hate the indeterministic nature of notion. Oh, sometimes you can right click, sometimes you can't. So, this is the new Apple intro. Oh, hang on. Got it. Uh, this is a new Apple intro. Looks kind of cool. Quite easy. Probably probably knock that out in half a day, right?

That's not Apple. That ain't Apple. They made this. This is practical. They call what they call practical effects. This was actually done. Now, I don't actually have the behind the scenes on any of this. Do you have the behind the scenes of how they did it? I No, but I will search it up. Uh, Apple BTS. Well, it's basically a load of glass panels, colored glass panels, and a lighting setup. Um, um, yes. So, I found it and here I'm adding it as the last link. Yeah, it's in the last link. Oh, should I share my screen? Yeah, let's let's do Yeah, that's fine. So this is what I found uh on the internet and we'll talk about this.

So I can't hear by the way. Should I share my thing? It's just a background music. There is nothing. Okay. Um yeah. So Oh, so it's How cool is this? So it's just using Oh, okay. Yeah. So they are doing multiple layers and they're editing those obviously uh on top of one another. Got it. So it's one it's one sheet. They're lighting it differently and then editing all of that together. It's not like multiple Exactly. So there is a lot of editing going on and some people here might will get into it think that why not do it in 3D why not do it with completely like with after effects because you could do it with any of these tools and you know you could try to do it with AI but doing it practically is different like it adds that layer of depth and that like details that you can add and that control that you can have uh that you can't do otherwise.

I guess with 3D doing it with something like Blender, you could set up an environment and create it, but I I do think like that the actual creativity comes together when you do something like this practically with real people. Uh and this is all about the creativity to me. Now, does this remind you of another logo that was created practically? Well, just before we move on, um I think also without being too again pessimistic about it, it also gives you this behind-the-scenes footage that you can talk about. And in the age of AI where everything is just created with a single prompt, there's no, you know, there's no there's none of this. Do you know what I mean?

That you can talk about it. You can show it's refreshing. It's it there's this it's more it's more about this it's more about the human connection storytelling than the actual intro itself. Do you know what I mean? So that's all I'll say there is that uh it it's for the behind the scenes but that that that's that's an emotional intentional decision otherwise they wouldn't film the behind the scenes. Right. Yeah. Yeah. Exactly. And I want to show something very much along the lines. Uh, have you seen the Microsoft Windows logo? Oh, yes, I do know. Yeah, the background. Yes. So, a lot of people might have thought that this was um also CGI. No, this is this is completely practical.

So, they have the like these I don't know windows or glasses and they project light but they they do so many different things. And here you can even see probably they are doing some sort of like 3D stuff maybe even uh but the actual thing is happening in in real and then they they introduce like actual lasers and actual smoke and the smoke and the lasers and like the light and imagine layerings. It's look this is it's crazy how they ended up because the the default background is none of this interesting stuff, right? It's uh it's just the boring thing. But so how can we see? There it is. Yeah, this is the final one.

So this is an actual picture. Then this is really really really cool. So Apple did this behind the scene and uh I had to make it about Framer again, but uh the the guy who works at Framer and he makes videos for Framer. Let me open up what he posted. He posted this and it's this is mostly trolling a bit and you know having fun with it but he made he remade the same thing with LOL and he you know with AI. So this is this is AI is it? Yeah obviously he literally taking the Apple video look this is like even the let's even compare it. Yeah. now. So, he must have put in that thing for training data then because it's it looks ex it's not like this uh similar.

It's the same look. But I think he you know there there's probably a lot of work that went into this. Did Andy create a behind the scenes because again people don't appreciate when something is AI slop versus when something is created with Well, this is my point. This is AI slop to me because this is literally taking somebody else's work, right, and saying like change it to, you know, change the the Apple logo to something else to change it to whatever. Um, well, he did lol to make fun of it, but uh how can I what's his what's his joke? Uh, okay. So, here he says, "You can't read this until I find the other link.

It feel it feels like this one might break the break containment. So I'll just say this about a few things. How easily people are influenced by a narrative feticization of real uh given the amount of V effects needed to actually get the result. So he's just saying why are we still doing this and why are we hanging on to this? But you know Yeah. And he al he's also saying like why people care about the real and to me it's about effort. We've been saying this for a long time. All I'll say I don't want to get too deep here to Andy if you'll ever watch this. Uh all I'll say is I got pretty my history with web flow is not so pretty.

Um I got addicted to being controversial to web flow for little to no reason. Right. It got views. I was like, "Okay, well, if I" And in fact, people found that refreshing about my honesty because I would look at that pessimistic side. All I'll say to Andy is don't fall in love with the controversy. Like the world is a much better place, needs better, you know, more positivity in it. We don't need negativity. Everyone's different. Yeah. The thing is like you um your stance was a stance of like honesty. you were honest about Web Flow and you kind of pushed it towards the limits, right? Uh the way you said it, you were you weren't lying.

It was just the way you said it, it was harsh. That's why you went viral, you know, the web flow u comments. Um anyway, this to me is a bit different. I'm not offended by this, but I don't like it. Uh that that's how I see it. And this person uh I don't know who he is, but he he made this comment and I kind of agree with this. So he says, "Congrats. You you have literally point blank proven that AI steals human work. You couldn't conceptualize this on your own." Well, I'm not like necessarily agreeing with this, but he didn't in this case. Andy didn't conceptualize this on his own. someone else created and executed the project just for you to download the video, feed into uh an AI video generator and type change the Apple logo to LOL.

They pay yourself uh then pay yourself on the back like you had any skill involved and um creating the final product. He meant patel on the back like pat yourself on the back. He's essentially saying this is stolen work. There is no creativity in it. And this is a bit mocking um the actual effort that went into the actual work. So my question is what side are you people are on? Should you should you just use AI to copy something in get inspiration or copy something or like go through the actual process like creative process of creating something from scratch. This is probably like not this but the one from web uh the one from Apple is inspired by something else.

I mean we have the Windows uh version from a decade ago right? This is literally from 10 years ago, right? So, all of these are compounding. Yeah. The AI I mean, all I'll say and this probably you never thought would leave my mouth, but all I'd say I think what Andy is has done is art, right? He's making a statement on feticization of real. I don't agree with his like why is that wrong? Why is that such a big deal? Why is it such a detriment on our um you know to to to love real things? Because you can you can love digital and real you know both can be the same, right? Just because people express a dis distaste for this specific implementation of something that's real and how great that it is real doesn't mean that they can't love AI and love what AI is capable of.

So what what Andy has done he he's making a statement here. it's less about the LOL and the the actual implementation. So, I will commend that aspect, but um yeah, I don't know whether his I go back to me saying don't fall in love with the controversy statement. I don't know whether his uh motivations for feeling this way are driven by his desire to cause controversy and and you know, yeah, we don't know that. But he's entitled to do whatever he wants. Do you know what I mean? Um, as with everyone, like I'm not too too upset about it. I just Yeah. I if if his art is not something that I agree with and that's that's why art is put out there, right?

To disagree and to to form your own conclusions and this that and the other. Anyway, art is an expression of how human feel. and if he feels about it this way by that definition in his art. But to me, uh I still care about the effort that goes into things and I think most people do. That's why most people don't like AI. So yeah. Yeah. Well, that's all we have time for. So make sure you follow us on Instagram and you YouTube. That's @ commandai show. Like, subscribe, and all that good stuff. and join us next week for all the news in AI design and dev. I've been Sam and I've been Kabarzo. Keep on vibing.

Peace out.