How to Create Ad Creatives With Open‑Source AI

This is what was used to generate one of those creatives that you're showing at the bottom.
Yeah. Yeah. Okay.
So, let's say let's say this one. So, how do you how do you use this? How do you generate those? So for instance, this one, this was an image and you run the image through a video generator which then animates it.
Mhm.
And then you stitch it into a movie or like a creative basically because all of these kind of a cuts that means that it's another image and another generation usually. So, in order to do these, for instance, this one already requires a little bit more um advanced workflow because Hi everyone, welcome to another episode of Aeroli 101. I'm Roman from Tangent and today I'm joined by Yakob from Two and a Half Gamers. Hi Jakob.
Hello there. Uh nice thanks for having

me. Uh yeah, Jakob from Two and a Half Gamers who don't know. What do you do y super quick intro for for people who might not know who you are?
Yeah, so currently I'm like 10 years plus in the game industry ma mainly mobile game industry. Lately I'm been working for last three years I guess as a independent consultant pretty much but uh of course yeah we run the two and a half gamers podcast with Felix and Mati which will be four years next month. So quite some time I guess.
Feels a lot longer, dude. It feels a lot longer. I'm not sure how you feel.
Yeah. Yeah. Yeah. Uh that's the grind there. But yeah, like I I work with multiple like gaming studios around the world or even non-gaming people these days because based on the whole like Dualingo, you know, apps pretty much taking over the app store. They're looking for our knowhow and like it's a perfect match a lot of times where
you know they need progressions, monetization and all these other things

like system design basically.
Yeah. Yeah. We've seen the same with apps like
huge amount but anyway uh we met at Moikto with Jakob and we decided that we want to talk about AI. Of course it's still 2025 so we have to talk about AI.
Um let's just jump in Jakob. It's going to be like free flow. Uh we don't have an agenda. We'll just uh see what the Yago has to show and I'll ask plenty of questions.
Yeah. Yeah. Uh there's lots of stuff and yeah, I guess this will be hopefully as practical as possible because this won't be one of those discussions that like AI will replace your job, AI will be this, AI will be that and so on so forth. This will be like what can you do now completely free and it's extremely impactful. So let's start there. Okay. So um I guess yeah for those listening best case scenario you can probably watch this on YouTube or somewhere there

because we'll be sharing the screen and I guess it'll be from now on some kind of a workflow. So, um, yeah, before we get to this nice image, which we'll get to in a second, let's first look at some of the actual stuff that's currently completely taking over the market, which is basically AI creatives. And, uh, AI creatives are actually the most impactful, let's say, surface level view of AI that we see in the market. is one of the most important things in the current environment because UA is more important than product in this year and next year even more and so on so forth like it was not like this few years before but now it is and if you want to give the best example just look at Kingshot Kingshot is the biggest game of this year it was launched somewhere like February and currently it's doing something like one and a half nearly 2 million a day and it's all based on this kind of a bait and switch fake ads, fake

on boarding, real gameplay, Forex style thing where it was actually taken from Thronefall, which was the the game on Steam. That was this was Wh There we go. Come on. That was pretty much very good, but again, very approachable. But what happens is basically they widen the funnel so much because it's so approachable. Users get to see these fake ads. Then they go into the game, they see the game play, which is as the one in the ads, which means like the fake ads, fake onboarding can equalize itself. Therefore, nothing's fake anymore. And it's exactly the thing that you've seen in the ads. But slowly the game unfolds you into Forex or some other high LTV engine that we see. It's proliferating also to other genres like social casino. Like just wait when we'll release the next episode on the channel. You'll see how this bait switch also works there. So uh and all of this is again possible because creatives and marketing is the key in this whole setup and AI

creatives I'm not saying like you can't do this without the creatives but it's enabling it very very big in a way that again it gives you volume because you need volume for this and AI creatives these days are extremely prevalent and we think that like our prediction is basically that by the end of 2026 there will around 50% of all create UA creatives either having AI hooks or completely done by AI like here you have an example the one that I showed before it was actually like a hook and there was the like creative real gameplay and so on this is the fully generated one where you would have stuff like like you see here completely generated in an image and video editor and that you just run it as your creative and that's it basically. So again, we won't talk about AI takes your job, AI do this, and we're literally talking about what's currently trending in the market now and how to get this. So if your creative team is

not using AI, you're already behind. That's basically the state of it. So how do we actually get to this and how these things are done and like a little bit more nitty-gritty stuff of generation? Because as I said, I won't talk about any other use cases about AI these days because in in my opinion, mastering the UI pipeline and mastering this and addition to boost your volume is the key. Of course, there are stuff like let's say you know it's just an example. Here's an example from from YouTube that I found where again you can use the confui thing which I'm using today and generate 3D assets through it. Again, open-source AI generation is not confined only to images and video. You can generate whatever you want basically in any modality as long as you have the open source model for it. The conf UI thing that I will be showing is just like the let's say the frame for it, but you can do from audio uh 3D assets, 2D assets, 2D sprites like you can generate whatever you want basically and

completely for free as I said as long as your graphics card is able to handle it. So that's there. So don't don't just think like oh yeah this is just image and videos and like it won't help us. though we can do pretty much everything because how I think the themes of the future will be going that they will be all making this custom because that's the biggest difference between the western approach of like blackbox AI tools which are again completely closed as for you can only do I don't know positive prompt negative prompt then like some very small customization to it whereas if we go actually to what we can do today. Um, yeah, it's kind of very heavy what you can do and what what you can actually create and check and stuff like that. It gives you completely free hands, uncomparable. And as I said, what I'm saying is that the teams of the future will be building their own tools and own data models and old data sets

that they will be then pretty much using through these open source AI models because that's the attitude or let's say that's the way the China handles it like China is currently flooding the market with all these open source models because it's their kind of also political policy of like we'll get these models in the people's hands therefore we control the ecosystem instead of the western
uh basically western uh approach which is like we have these giant open AI companies that are doing like the best of everything but again it's not that supportive as here in China is in China the community is also driving these models because they're adding all of these like additions and stuff imagine it basically like Skyrim like Skyrim is played to this day as one of the best RPGs in the world why because it has a giant modding community that revives it patches it improves it so on so forth so that's their approach basically That was the f your first creative when we started had the Chinese characters and I already because I also follow the channel. I I know that you have like

some folks from China and they're like sharing some like crazy stuff and that was my first question. Do you feel like they're up front than like everyone else with this AI adoption? And like clearly you're saying yes, right? I I would say so like uh not only their models are again they're open source so you can go customize and use them for yourself but the approach and pipeline is again different in China because again this is the big difference between the west and the east user acquisition is the most important job in mobile game industry in China
in the west it's not in the west it's product usually product as for like either design or like you know live ops PM monetization stuff like that. That's the most important part in the core of it. User acquisition for them is again as I said the most important part because also the product is so up to par across the whole kind of industry there.

So their product is great to begin with but yeah that's another discussion for some different
but can the folks from west adopt this kind of like the models are open source as you said. Yeah, again they can like you know we have AIs all over the place so there's basically no language barrier if you know how to use them. It's just artificial like it's it's like
you know effort
based like you need to put in some effort and then you have it
but other [snorts] than that like yeah it's it's it's quite easy like I can do it. I'm not a programmer like I'm a game designer. I can do Excel sheets with like maths and economy but I can code and I was able to do all these things. So it's not that hard. Uh yeah, everybody can do that and it's again just people in the west kind of sleeping on themselves whereas they should be doing these things all over the place. But yeah, we'll get to it. Um, yeah. So, as I said, how to do these creatives and like how to pretty much even get to some

of these things because again, you can do and like do this still pretty easily through like Nano Banana or CH GPT or any other image generator in the west. You can still do great like don't get me wrong. This is more hardcore and like uh let's say more customizable stuff because like what what can you do and what can you create? You can for instance create your own LoRa. We'll get to it. What does it mean? But basically what it means is that you create your own data set from your art, your custom art, your whatever you want to do and you add it in onto a model. Therefore, the model suddenly spits out like an art that would be coming from your artist, which isn't really the thing that you can do with GPT or or these other tools because currently as I'm seeing it, for instance, every big company and I mean like companies like I don't know Blizzard, City Project Red and all these other guys, they're probably already creating their own models which are completely fed only on their own data. meaning that they're again like creating

the armies of these artists that they'll suddenly be able to do any use which is completely legally okay because there's no copyright as for the you know they're just using the model not the training material but yeah that's again one of these things so how how does it look and what's there so this is comfy UI let's start maybe from a little bit easier workflow until we get to the hard stuff um again it's quite easy is visual prompting once as you get into it. So you just download the thing from hugging phase. Hugging phase is the big uh programmer repo with all the databases and and models and everything. It's all open source on the internet.
And the important part like you're looking at this like oh this is so like how did you create? No, you don't. You don't need to. It's very easy because all of these things that you see here for instance these workflows that I have here you just take from someone else. Like if you hardcore, you can literally go and like okay add a node and like
edge spaghetti here and like do this visual coding thing that you can you know goes from here from here from here.

You can do it yourself but I don't because like you just for instance [snorts] this one that I have here um the big one. Yeah, no chance for me. But again what you do you go on the internet you read the guide and on the guide you have like this whole thing pretty much and again somebody do it for you. So, don't get maybe let's get rid of this so it's a little bit more uh easy on the eyes. Uh don't get scared and don't think like, oh, this is just like horrible. As I said, I went through these. I didn't know about all of this. And pretty much by trial and error, you can figure it out quite fast. It's not that hard. And my number one advice when working with these tools, whatever errors or like stuff that you have there, just throw it into check GPT and it will just tell you like in layman's terms like you need to do this, you need to do that, you need to do this. And it's great because again we need to realize that suddenly we have this AI that's literally right there sitting in the corner for us which we

can ask anything and it will do anything for us. So all of these things like I don't understand this, I don't understand that doesn't matter because again you slap it into AI it will tell you and especially programming code immediately it'll fix errors and do stuff for you. So it's again effort based barrier no other barrier. [snorts] So um if we go into the basics
clarify uh maybe for the small one uh this is what was used to generate one of [snorts] those creatives that you're showing at the bottom.
Yeah. Yeah. So let's say let's say this one. So how do you how do you use this? How do you generate those? So for instance this one this was an image and you run the image through a video generator which then animates it
and then you stitch it into a movie or like a creative basically because all of these kind of a cuts that means that there's another image and another generation usually. So in order to do

these for instance this one already requires a little bit more um advanced workflow because one thing that we have here is a consistent character.
Yeah
which is like yeah it's not something that you see every day and like yeah so again for this you use conf UI where you have workflows for consistent character literally create a character and from that point on you kind of save it like this is my character and then all the generations can go through that character. Therefore, you end up with something like this where I said like, okay, let my character sit in the evening in the office and there it goes. And the video generator is just kind of a cherry on top. It's not that hard. It's just like the most important part of the whole image or let's say creative video generation is actually the image itself. Because the workflow that you go always is image to video, not text to video. M
lots of times people just go text to video like you go to an image generator and you do something and like just input some text and it just generates

something which is great but you don't have control. That's the big problem. You don't have control how it looks, how the characters look, how the environment like anything looks. So again the key to video generation anything is image generation. That that's that's the number one rule that that you learn with these things. Therefore if you want to have great creatives you first needs to master the image generation. Once you master the image generation, then always the first frame starts with your image and from that image you go and create the creative and you can do pretty much whatever you want. So how do we get to the image generation? So as I said you install stuff like confi you can do nano banana or whatever like stuff there anything is good but this is just much better in a way of controllability. So let's just go over this like very simple workflow and like uh how it works and what what we have here. So this is the Z Z image Turbo which is the latest model from Alibaba that is literally taking over the internet in last month for

those who don't know. Uh it's unheard of because this is very small model literally like six billion one parameters and it's outstandingly good but yeah I'll just go very fast through it. So here for instance we have the base model which is quantized. Quantized means that in order for these some of these models like we don't really have the I don't have the top-of-the-line graphics cards. So the community again creates lower versions of these models to cut down on the VRAM requirement but also a little bit on the quality. So that means that I can run this on my 380 Ti GPU graphics card which has 12 GB of VRAM even though the base version of this model requires 16. So you literally go on the internet and again in the guide itself like I have here for instance that again you can get and like find. So you have these repositories in again these are for instance the

quantized version of the model is go all the way into like small one which is like two gigs or whatever and you can run this even on six gigs va card. Uh
so the first step is actually to buy a good computer. Is it is it what it [laughter] is?
Yes you need a good computer the past when like it was important to have a good
Yeah. So we need we need at least something like I would say 8 to 10 GB of VRAM Nvidia GPU like this stuff won't work on AMD maybe in some experimental form it it will but you need a CUDA core GPU that's the first step once you have this you need confi UI again you can get it on the internet very easy it's just one repository from hugging face uh also I recommend installing confui manager which is just the UI add-ons stuff pretty much utility that you again doesn't need to go on the website download manually. You can just click in it and like it will download it uh from git immediately.
And once you have this again you just drag stuff and you literally can like go

here and drag an image here and the image and its metadata will then create the workflow if it's embedded in it. So that that's the beauty of it like you don't really need to create all this spaghetti visual coding stuff. It will just have the like for instance this one is example workflow on the site. It was just like throw in an image here we go. So again what we have here and what are like some of the things that you can control here and what gives you the things. So here [snorts] we have the base model as I said uh the text encoder and the model itself it's quantized so it's lower quality lower VRM so we can actually run it. Then we have stuff like shift this is specific for the model is more of like a contrast slider. So less shift means less more contrast more shift means less contrast. That's there. Then we have the positive prompt. Um yeah, I'll get to it how I got it. And the negative prompt if I understand this one doesn't really work with negative prompt that much like it's again uh some

some image generators don't even have that like flux for instance they don't have negative prompt. Then we have the image size which is just like a square of 1,024 bits times the same. We could pump it up to 2K even higher. Uh the problem is that it will just load longer and we don't need it for the sake of this video. So that's there.
Jakob Jakob. Uh quick question. Is it also like effort based as you said at the beginning in order to understand everything you actually
Yeah. As I said, no programming skills on my side, no computer science, no nothing. My background is psychology.
Like you don't need anything. You can get these things still like as I said for instance we can link the literally the how to guide tutorial into the video. There's like a 40-minute tutorial but most of the stuff it's not even a tutorial. It's just the guy goes over like what's the what's the comparison between these models uh let's say here for instance. So we have zimage flux and

quen is more of a comparison. So really the like where he goes through file manager and like just tells you how to install it. This takes like 10 minutes honestly. Like it's not like it will do this and that and it will be super hard. No, it won't. It will be just like four or five clicks. Again, you have Chad GPT sitting right next to you that if you don't understand, you just tell it. I don't understand this. What should I do? It will tell you. It's that easy. Like for instance, I didn't understand like which which quantized model I should pick for my graphics card. And uh yeah, so this is what it told me. So I just literally pasted the repository uh from the thing and it told me like okay so you go here and these are the models. So if you have 10 to 12 GB VRAM pick this one because this one will probably be enough for your memory. That's it. And you do all these steps like this. It's it's super easy.
So nothing really to it.
So once we have all these fixed let's just finish the last step. So steps are very important. This is the setup that

tells you how many actual kind of a parts of like generations it goes through because all these images usually the diffusion models it starts from noise. So it's like imagine just black and white grainy picture that like all these pictures starts like that and this will be like how many number of steps like the noise will be run through this. Then we have CFG value which is how much prompt adherence compared to creativity we let the model do meaning like how much more creative we let it be compared to how like it must be exactly as we prompted like again a value that you can play with and then some like base stuff that like you don't really need there. So if we go go here and run it, we have this kind of a demon guy which is like hyper realistic line of driving furious forest spirit da da da da da da. Let's run it. Uh we have the same seat. Yeah, we need to change the seat to random

because we don't want to have Yeah. a different seat each time.
Prompt. Mhm.
I see a lot of text in there. Is it?
Yeah. How do we get this? Yeah.
Yeah.
Exactly. Yeah. Let me just generate the thing. So you see it went through the prompt. Now it's in the K sampler. And then from K sampler, it's go to decode. And then there we have our image. So we have instead of this guy, we have this guy. That that's like quite easy.
Uh how do we get this giant prompt? So prompting is kind of Yeah. another way of like learning these things. Um, so for instance, this prompt I got from CVI AI. CV AI again is one of those things that I would recommend you go check it out. It's pretty much uh the biggest open-sourced community website on the internet. Think of it literally as an

Instagram. So it's just basically images and videos of other creators that people vote on and then like can check and do stuff. The very important part about this site is that you can go there and learn and get stuff for yourself. So for instance, our forest spirit is just I was just browsing here like for instance the everyone the today's images like what's that generated you have some very interesting stuff that you can get here. Uh, by the way, spoiler alert, I'm using the CV green site because there's also the CV AI comp site, which is like 90% porn because that's what people generate with user generated context. So, just saying if you want the one without it, it's the green one. If you want the one with it, it's the base one.
Thanks for picking the right one for [laughter] this recording. I appreciate it.
No worries. So [snorts] again I just found the image from a creator and the key part here is not the image itself but again this thing on the right which we can zoom on

a little bit. So what we have here is that it tells us actually how this was created and we can even run it on the site itself and generate it there if you really want. The site allows it if you if you buy literally through credits but again why should we do it if we have it open source. So what this tells us it's using the zit image tour generator. So I can literally just go here, click here and then I have the model. It was released November 26 and I can download it or create with it or basically get stuff from it. You have also some kind of um current uh generations and like what people are doing there and stuff like that. But again we already know the model. Then we have the prompt. So we have the prompt. We can take the prompt and you can play with it and use it. Prompts have very specific setups. Like again, we would need probably a different podcast for it. But again, you don't need to create this stuff yourself from scratch. You can learn from other people. This is why this site is so

important comes into the formula because you can create amazing stuff just by copying other people's work and reverse engineering it and seeing how it works and therefore you learn and you learn very very fast. Then we have also some other important things which is the metadata like basically how the guy specified his sliders in confi. So we see as we talk cfg scale a little bit more to the adherence. So it's 1.1 only eight steps the sampler we can even take the same seed and generate the same exact image if we want like that's also possible because he left the seed here. Some people don't share their generation metadata because they're like very you know want to like stay confidential and stuff like that because some some people work like very hard workflow but most of the stuff that you see here you can do and you can just take and learn from it. This is the beauty of the site that you learn so much. So again this was pretty

easy to do and we can do whatever we want actually like just for the sake of it. Uh, so if we go here, we can leave our fire guy and Yeah. Roman, tell me what do you want to generate?
Uh, let's do something Christmas related.
Mhm. So Christmas what?
Zombie.
Zombie.
Like, do you remember plants versus zombie?
Okay.
This is what like immediately sparked me,
by the way. Good that you're saying it. Uh the beauty of these models um Christmas postcard. Yeah, let's try this one. The beauty of this models good that you mentioned is that they're completely uncensored
which is again the big advantage of it because if you go into chat GPT or again one of these kind of main models you can generate IP based stuff like for instance my son asked me like oh can you

can like can I have like Olaf or whoever from Frozen or like no you can't because these models have other AIs that are sensoring the output of them so that you can't do it like it's impossible here.
Quick.
Yeah, it's very quick. Again, as I said, I'm using a downward quality one. So, this would be a little bit different one uh than than the usual quality that like you can pump it up and there are still better models. This is the turbo one. So, the speed is more important than quality itself. But again, whatever you do here, you see you still can get like amazing quality. But again, if I would go and as I said, uh, Ela and Anna from Frozen standing in front [snorts] of a giant frozen castle, cinematic, high quality,

realistic. Let's try. Um, yeah, the more these tags and words you add to it, the better the image will be. Of course, that that's like without saying. Uh, as I said, I would recommend for anyone to learn just the process. Oh, there we go. See?
Oh, that's literally Well,
yeah.
Yeah. Like 95%. It's like [laughter]
if we would fine-tune it a little bit more with like details and like you see the ice maybe needs a little bit stuff like that here and there and like yeah we can get to it very easy.
Your legal department is not going to be happy about [laughter]
Yeah. Yeah. Yeah. But again you can do whatever you want. That's the beauty of it. So it gives you and it's completely free, you know, just takes electricity and your GPU. Nothing really to it. But again uh I would recommend for anyone just to kind of touch this run through it and just like learn it because again you can apply this same process like how

this works to any modality to like as I said text to video, image to video, 2D art, 3D art, voice, you know, whatever like it works the same. And I think it's important for people to understand like what's under the hood and how much control they can actually have because it's like amazing. And we'll probably end up with this last thing which is my signature stuff that I was working on. And uh yeah, this gives you much much much more control. This is a very advanced uh workflow that um not this one. Sorry. This one. There we go. Let it run because this one is actually 240 steps. What does it do? I I [laughter] didn't understand. What does it do? Yeah. Yeah. So what we have here is that we are actually using an ion on justice anime model and we are using the model only for 140

steps. And what we're trying to achieve we're trying to generate uh snow leopard anthropomorphic warrior in realistic style for our game. There's pretty big prompt here. uh pretty big negative promo also like to took some time for me to do this but we want this to be realistic and the anime model that I'm using it here is not able to do realistic stuff so like what's happening here so what you do you use a refiner so what it does after 140 steps this model stops and I actually plug in a different model so now we're like doing two model generation now through phenodin which is a realistic model which finishes the generation the d the noising of the noise from the image for another 100 steps. So it goes all the way to 240. That's why it's taking like 3 minutes. And then it basically creates something that the each of these models

couldn't create on their own because we want again a fantasy style snow leopard warrior guy that that like again I was not satisfied with anything I found on the internet. So I just dig deeper and deeper and deeper and deeper and like got into it. The very important thing is that this model and the workflow that we have here and by the way everything that you see here we not using even half of it like here we have basically the the possibilities of this workflow and you can just plug them out like functions like you know just click here and like enables it or no all the violet stuff that you see it means that that's inactive like we're not using a open pose IP adapter or control net upscaler all these other things like it can do so many things that again would take like different podcast to do. But uh what it can do is still we're using the after generation corrections like detail. This is the really important part because in the image that we generated here for instance you see that like yeah they're

great but there's something strange about these two like it's not that you know the position of their eyes and like everything. It's like like they look from wish. So what happens here is we can look at in the real time actually as the workflow is continuing and okay it's already on IDTaylor. So we have the base image here and you see it's not perfect like it's like the face kind of distorted like yeah we don't really want this. So what's happening we have a face detailer and the face detailer actually fixes only the face. So, we are putting another generation on the image that we already have here. And not only that, we also fixing the eyes to make them a little bit more better.
Oh, yeah. Yeah. Yeah. I see. I see.
Basically, and you can do again there are like four passes. We can do both hand and body kind of a setup. You see how the body is kind of again fixed a

little bit. So
yeah,
last time I I was checking some stuff on the internet, the like a professional from AI like agency that that was like sharing his workflow said that it takes him something like 20 hours on an image and 500 generations to kind of get it where it wants to be like a top quality. So just to give you an example like from the really really basic stuff like let me generate Olaf from my son to like very very advanced stuff like this is how it works because again this is something like that needs to be kind of perfect because it again defines what you want to do. And if we go again somewhere here, uh not this one, but um the one creative that I got really really um not this one. Yeah, there we go. So you see how beautiful these creatives

are. Literally like Pixar movie. And again, you get to this quality by being able to use advanced workflow. And what you end up with are these perfect creatives afterwards. So again, that's the beauty of it because this looks literally like a highle cinematic. It's like something that somebody would take again lots and lots and lots of work and time to kind of get and generate. Uh I mean draw, but then again you can just generate it through pretty much like advanced workflow timeline and yeah it would go and you need consistent characters and all these other things. But as I said it's like step one to getting all these things. So for any creative team that are making creatives uh yeah I think AI is the image generation and video generation mastery is like existential problem next year basically because you just won't be able to keep up with the volume. It's just like very very hard and like having enough creative volume means that your CPI is

getting low as it should be and it's getting higher. So eventually all of these things translate basically into that. And as I said, even though this looks like super complex and stuff, it's not like I'm not even scratching the surface of it like how complex it can get. It's just just some basic stuff that I'm showing pretty much not not really to it.
Yeah. I really like how we look at both of the schemas like we at the end of the day we generate an image.
Yeah. Yeah. Yeah. Yeah. Right.
Yeah. We have a nice warrior here. But again, uh you can do whatever you want in the end. And that's again, as I said, that's the be you can play with it and like try it for yourself. There's not some like money hungry website consuming credits or whatever. Like as I said, the only thing that you need is your GPU and electricity. So you can do whatever. Like I'm currently myself past like something like
12k 13k generations probably. So
yeah, sometimes like even here for

instance, there's a setting that like run it in batches of like eight or whatever. Like you can just go and sleep and like let it run and then pick the best one. Like I always do that sometimes like uh yeah, it's like an idol game. So you know, you come back and you collect your rewards basically after it. So yeah, but but I think it's very fulfilling in order to be able to like know how this stuff works because then also what it gives you is like once you go and once you see these creatives and all these other things that are currently trending on the like most of the times I can even spot the generation model
just by looking at it because some models have like very specific kind of a you know you can see the giveaway like for instance all the Ched GP image. They have this orange tint behind them.
So if we go I think like one of the last ones here.
Yeah. See this one?
So this is a uh creative that Candy Crush is running and you can see it in

the end that like there's this kind of orange aroma around the image which means that it was based on CH GPT as the base image. So here here it's kind of very very very visible.
Yeah. But again it's very subtle but once you see 10 thousands of these images like you you like it's very obvious afterwards.
So therefore you can immediately see how these teams are working how they are generating these things and you can do it yourself. again gives you a lot of edge over the process
and like it's really a skill right for probably 2026 for if you're doing uh creatives whether you're like a UA manager and that's part of your responsibility or you're part of the creative team.
Yeah. Um and just trying to summarize at the end and the cloud-based services like chat GPD cloud will give you less flexibility with what you can do.

Therefore, we would like to use open source models.
Yeah. Yeah. like like honestly you can even combine those in a way that like for instance you can do the image in your open source model to kind of refine it better and then use like [snorts] a video generator like video generator is quite important but it's like the final part of the thing where like you know what you want to generate which starts from the base image which can be com combination of like I don't know use the confu I want to generate the image and then like v3 for like finishing you know the video and stuff like that again you can do whatever you want best case scenario you test all of it and you figure out which one works best for you. Like that's that's the beauty of it. But again, by knowing these things, we even have the option to test it yourself. Otherwise, you just like, oh, we can just use V3. That that's the only thing we know and and that's it.
I see.
And and and the other big problem with these like these things degrade very fast and not not really degrade, but pretty much new stuff gets released all the time and you need to keep up. Like I

I've seen some of the creatives here uh for instance I think yeah these ones or maybe a little bit older ones that you can see like some of those are just running on old models like people just haven't updated yet which is again like normal because this was like the update cycle here is like 3 months or something but you want to be using the cutting edge stuff because again gives you edge on like quality stuff like that and all these other things. So yeah, just kind of moves very very fast.
How do you keep up Jakob? You personally like do you have time? Uh I don't know.
Okay. Yeah, understandable. Um so I listen to few podcasts of course uh based on AI literally we can link uh the this one is like in my opinion the this kind of AI search YouTube channel is literally a guy that just do the news of like every week he goes like what was released this week and just like goes through the all the models and they have

specific videos on like specific stuff of like the comparisons stuff like that like so this one kind of a debrief I watch very regularly Then I have few podcasts that are like industry based of like what's the latest of CH GPT versus Microsoft and all these other things. And then of course you go on the CV AI site and you just go and see what people generate. So for instance here you could clearly see that lots of these things like okay there's a lot of CH GPT actually in it. There's a lot of uh what's this? Yeah, Google Nano Banana pretty much is trending really high these days and you just see like what what people generate and what what what's pretty much there in the market and this just tells you like okay like yeah that that's basically it. So you know who generates what
right and like this is so interesting that you're actually like game design expert and you are now into this like creative part of the game like full in. How does it feel Jakob?

like [laughter]
oh it's it's great you know like I'm always kind of obsessed in like but again the biggest problem was I didn't understand how this thing works which for me is the biggest itch of like I need to do something about it because I don't feel safe or like how do we say it I don't feel on top of it if I don't understand how it works so it kind of drives me to kind of go into this kind of rabbit hole and learn about this because again if you don't know something at least know how it works you don't need specifically do it, but at least know that there are these options because that that way you won't get sidelined. You won't get in situation that somebody tells you something, you can't call their and you don't know if like this is the best, not the best or are they even telling you the truth, stuff like that. So again, and AI is just moving so fast these days in a way and it's like one of the most important technologies of our lives currently. So why not you know slap two birds in one stone where again we need this because of our game industry professional expertise and on the other

hand of course like AI will be there and like it will change stuff that that's for sure but again the edge that it gives me now is I know for instance like gaming wise no it like gameplay wise it won't change stuff that much like it will maybe help with some optimizations like matchmaking or whatever bolts I don't know what but we still haven't reached the point that like AI is doing the you know the AI enabled games like games that won't be working without AI capabilities that we we still still didn't hit that inflection point. It's not something like it was for instance in I don't know 199 something where Doom 3D was released because the 3Dness of it enables you to do stuff that you couldn't do in 2D. We haven't reached this point yet again. Why? Because you learn about this and you know that like oh this doesn't make sense. It cannot even code properly yet. So much hallucinations.
Everything is connected.
Yeah.
All right, Jacob. This was like super insightful. I'm sure the guys will have

a lot of questions. I'll ask everyone to leave their questions in the comments. We'll ask Jakob to answer them when once he has time. uh any parting thoughts or we will of course leave all the links in the descriptions to two and a half gamers and the stuff that we mentioned during the videos but I also want to give uh Yakob to say something especially at the end of the year. Um last parting thoughts from you Jak. Yeah. Yeah. Um as I said if you have any questions or any thoughts comments uh feel free to leave them under video or like you can join uh the two and a half gamers slack that there's also open for all all the people to kind of share their knowledge and talk with others. Uh yeah I would as I said like parting line is go and learn it. Don't wait for it until it will kind of you know it's too late to kind of catch up. Well said. Well said. We'll end on this point. Like and subscribe. I'm sure you like this episode and thanks a lot,

Diego.
Yeah, no worries. See you there.
Cheers.
Bye-bye.