Truth be told, I try to space out my AI discussions on this podcast, but I can't help but talk about OpenAI's latest image generation update. So by now, I'm sure you have heard all about OpenAI's new image generation. And just a quick recap to get you up to speed in case you're not already. You've always been able to create images in Chatubit. It used something called DALL E, but it was never really that good compared to what else is available in the marketplace. But now with this latest four O update, it is the best. Not only do the images look absolutely amazing, the groundbreaking, technology here is really like two things. Number one, it's the ability to incorporate text. Before, the text would just be blurbs or it would get, like, a couple words right. Now it gets a lot of the text right. And that makes a huge difference when you're creating things like a soda can or a poster. And we'll talk about some of the use cases here in a second. But another reason why it's such a big deal is it can intuit better. So if you give it a prompt, say you want to turn something into a comic in a certain style, oftentimes with these image generators, like Midjourney, Grok, or how DALL E previously was, it might get close. It felt like maybe 60 to 70% of the time you'd get kind of close of nailing what the look is that you're going for. And now it's just so much closer. So you can give it a little bit, and it can kinda fill in the gaps of what it is that you want. It's kind of like how I was talking on the last episode about how viewing AI is your assistant. It's like your assistant got smarter. And just to cover some of the ground bases here in case, you know, I don't wanna assume anyone knows everything. So who has access? Right now, everyone has access to this image generation. As of just a couple days ago, they opened it up to everybody. If you're on the free plan of ChatTBT, you only get a few image generations per day, and that is increased a bit more if you're on the plus plan, the$20 a month plan. And then that's increased a lot more on the pro plan, which is $200 a month. Now I'm on the$200 a month Pro plan, and the first couple days after they released this image generation update, I had hit some limit. It was like, you'll have to wait four hours, but I haven't hit that limit again. So I don't know where the limit is with that. I don't know if there still is a limit for Pro. I would imagine there is. And I don't know the number or limit for Plus, but I do know that there are two places you can access it. Number one, straight in ChatGPT. So you ask ChatGPT, you say, hey. Make an image of blank. And then it will create it right there in the chat thread. Another place you can access it, which is also my more preferred place, is using Sora, which is just sora.com, s 0 r a Com. And there, it's a little bit more dedicated. So you don't have a chat bot back and forth that you're creating the image with. You just input the prompt, any reference images, and then you could select the dimensions and how many versions you want it to output at a time. The more you select output, the longer it's gonna take. So you could put in a prompt and a reference image if you wanted, and then you could say, you know, you wanted it in, a square format dimension and you want two versions of it. And those two versions will look different, so you can kinda pick and choose. Now I think, at least on the Pro plan, I think you can have up to five generations running at once. So, technically, that would be like, what, 20 photos being generated at once, which is quite a bit. But, again, that's the pro plan. I'm not super familiar with the plus plan. And I wanted to share some interesting use cases that I've been using this new image generation update for and then what I've seen other people using it for as well because there's some really interesting stuff happening. So I've been using it for creating comics. You might have seen on my Instagram. I've, like, turned myself into a comic, and I have, like, a little bit of a backlog because I don't wanna just be, like, generating comics all the time now. But I've been making these comics on these ideas. So now I'm thinking through the lens of, like, everything in four panels. And I don't just completely outsource the creation to ChaChaPT. Oftentimes, I have to refine things. So you might see an image and you're like, oh, did ChaChaPT make that? And it's like, yes, it did. But I might have, like, had to hodgepodge some things together. I might have had to take it into Canva or Photoshop to fix something. And the ideas and, like, the text on there is usually something I made. Like, the whole premise of it is what I made. So, yes, while you see it and you might be like, oh, did Chad Gbitt make that? It might have actually taken me, like, two to three hours, if not longer, to come up with the whole thing. So if you see something that I made, don't think this was, like, a one shot generation, when it comes to the comics that I just used the first thing that Chatuchiputti made. But I wouldn't have been able to do it without Chatuchiputti. So it's worth saying, like, hey. I'm not doing the heavy lifting. I cannot draw very well at all, and Chatuchiputti is doing the heavy lifting. I'm just doing the ideation around, like, the angles of the comic and making sure everything looks like how I want it to look. Another thing I've been using it for is companion images to post. A little bit less involved, but, like, on LinkedIn a couple days ago, I posted about how I am gearing up to hike Mount Whitney. And I was like, oh, you know what? I should just throw in a comic of me on that. And it was just a simple comic, me hiking Mount Whitney. So it's cool to put like a little companion image to post. Or in my Instagram stories, I replaced me, lying on stoopy's house. Just a fun little post. I've also been doing some logo idea generation specifically for content clips, and I would put in this prompt, coming up with this, like, pop art style just to get some ideas because I'm always thinking about branding and design and stuff like that in the background. And I got some awesome ideas. In fact, I got one idea that was, like, blowing me away. And I was like, should I change the branding? And I'm like, you know what? I came at the conclusion. I should just leave the branding that I already have. It's not that important. But if I did wanna change it, I could use that. And I could see how that would be a great use case for someone else. We'll talk about that more in a second. And then also, I've had a lot of fun turning family photos into, like, themed photos. So I'll take family photos. I'll turn them into, like, Peanuts style because my daughter loves Snoopy or Sesame Street style because my daughter also loves Sesame Street. So I'll take photos of us, just normal family photos, and then turn it into that. And she loves seeing them. And I can't imagine what that's like to my kid's brain. Hopefully, that's not bad. But you you know what I mean? Like, we didn't obviously have this stuff when we were kids, and it wasn't as good as it is today. We didn't have anything like this. Right? I remember learning on, like, a black and white Macintosh computer. And here I am with, like, my iPhone, and I'm on a Ferris wheel with my kid. And then, like, you know, two minutes later, we look like we're in, like, the Peanuts universe. It's pretty wild. And some things I've been seeing other people post about or share examples, things like that, just to get your wheels spinning around what you could use it for, is colorizing photos. So say you had, like, a black and white old photo, feed it through ChatGPT, ask it to colorize it while keeping everything else the same. Also creating mock ups for websites, so, like, homepage design. If you're thinking, like, I don't know what the homepage should look like. Well, now that it's so good at doing text and design, you can get some ideas. You might be thinking like, well, I mean, that's great on paper, but like, how do I get that to the website? Maybe that's coming before we know it, but that's also a limitation too. So while you have this awesome looking thing, maybe the thing is you just find someone who can implement that. Because before we would like go to the person making the website and be like, you know, I'm expecting you to design it, to build it, the whole thing. And that's really expensive. And I know for a lot of people listening, they don't have the type of budget. So if you could handle at least the design part of it and then pass it off to someone who knows how to build it, which are two different roles, that is pretty dang cool. And, again, you know, I could just, like, nerd out on this stuff all day, but you could incorporate the best practices. So you could start out your chat GPT thread by saying, like, you know, what are the best practices for a personal website? And it will say, like, it should incorporate these things, these things, these things, and you're like, okay, cool. That sounds pretty good. Then you go, can you make me a mockup for what the homepage should look like? And then you could go there, you know, just these nuanced things. Again, not taking the first thing it gives you, but having something that didn't exist before you started getting the ball rolling. And if you ever listen to interviews with, like, Sam Altman or these other AI tech CEOs, a lot of them talk about how we're just gonna be going faster from idea to implementation. That's a lot of what I see happening here is you have an idea, or maybe you did invent an idea, but you had an inkling of an idea and then turned it into a bigger idea and you're able to implement that faster. Now sometimes you can implement them by yourself and sometimes you need to get help. Another thing I've been seeing people, post about is obviously just creating profile pictures, also creating YouTube thumbnails. You can extract a thumbnail that performed well for somebody else and then use that as inspiration to creating your own. Now there's a whole ethical debate in all of this, which I'm not gonna get into here, but I'd rather just, like, fill you with ideas on what I'm seeing than sit here and talk about each one and, like, whether it's ethical or not. I've also seen people create infographics, so, like, recipes. Say you got a recipe for making pasta. Well, now, Chatapiti can take that recipe and make it into a great looking image that has it laid out step one through five of the whole recipe with images and the text and everything. Also creating textures. So say you had a hoodie or something like that, and you wanted to just pull the texture of the hoodie, it will create this close-up image. And you can use that in all sorts of different ways, kinda similar to, like, also creating backgrounds, maybe for images or videos you're doing. If you wanted, like, a specific look of your background, you could have ChatChippyD create that now. You might have seen my video. I posted it on TikTok and YouTube Shorts about creating podcast show art. It makes fantastic show art. If you did see that video, the example I used in the video was the very first image that it generated. So that wasn't, like, a lot of back and forth. I also shared the prompt in that video, which was just very simply telling ChatCha BT the name of my podcast, and I gave it a reference image to work with. Also, I've seen people create book covers. I created a book cover. I was showing a friend the image generation update, and she's in the process of writing an ebook. And I was like, oh, check this out. Okay. Let's use your example. She told me the name of the ebook she's making. We made a book cover, and she's like, ah, I don't like that. I'd rather have it like this. So what did I do? I just literally as she's saying, then I go, okay. Type that in. Press enter. What do you know? In, like, two minutes, there's a fixed version. She's like, that's amazing. Also creating stock photos. The people that it's generating are, like, indistinguishable from real people at this point. Also and this is, again, not talking about the ethics in aid of any of this stuff, but creating fake receipts and screenshots. So I've seen people create these screenshots that is like, I've earned$10,000,000 and it, like, has, like, the name scratched out. It looks like a legit screenshot taken from, like, their Stripe dashboard, and it has their name on it. The whole thing, like, you really wouldn't know that it's not real, but it's not. It was just created in Chachapiti. And similar thing for receipts. People say like, oh, I paid x amount for this thing, and it's not real. They were just using them as examples of, like, what could be done. Of course, not ethical whatsoever, but it's worth sharing these interesting use cases. So what's some things we should keep in mind when it comes to this big image generation update? Like, this is a huge upgrade and it raises the bar. So something I'm keeping in mind is this creates a new reference point of what these other companies are gonna have to aim for. Like, they can't come out with just better looking, more realistic people at this point. They're gonna have to incorporate text. They're gonna have to get better at their models being able to intuit what the person is asking for. And it's only gonna get better, faster, more capable. Like, think video. Right now, we're thinking images. There are these services like Runway that can incorporate video, but, like, just keep thinking bigger. We're looking at really great looking images with text. AI video is not quite there, but think about how fast it's going. I mean, it's been, what, two or three years since ChechiPT has been out. And if you think about when humans went from drawing painting pictures to photos, to black and white photos that, you know, only people would have to travel in from out of town to do. Then cameras became more common. Then you got color photos. Then you got motion pictures. Then you got everyone with their handheld cameras where you're taping over things and you're losing memories because you're like, I don't have any more tapes to record this thing. To DVDs, to MP4s, and being able to just record everything from your phone and storage really becoming a commodity and you're not having to, like, worry about it as much these days. I think the same change is happening with AI pictures and video just at a faster scale, and it's just gonna become more and more capable, which brings me to another change in our relationship with media. We have to be more weary and conscious about questioning what we see. These tools are fun. They're powerful. They're a little unsettling. Okay. Sometimes a lot unsettling. But when things clear, we're not gonna go back and these aren't gonna get worse. So the real challenge, I think, is gonna be staying grounded while the ground itself keeps shifting. And there's a lot of people who are pushing back against AI generated stuff, and I think it's just a losing battle, honestly. I think the more we can lean into it, the more we can accept that it's here, the more we can work with it in a way that is productive, the more we can harness and the sound's AI generated. The more we can harness our own power for using these tools. And I think there's, like, one example I'll end on. There's, you know, someone, I would say like myself, who will put out AI generated stuff, and I don't get a lot of pushback. Like, no one's like, stop with this AI stuff. Like, I don't hear people say that. If anything, people are really curious about it. They may be a little bit intimidated by it. And some people are, like, ashamed that they use it. And then I've seen other people where they share AI generated stuff and the pushback is so loud. And they're like, you could have hired a human to do this. You could have done that. And the person that shares that, like, doesn't have the budget or the intention to hire somebody for that thing that they made. And so it wasn't like, oh, well, we could have got someone to do this. So we went the AI generated route. It's like they maybe just wanted to hop on trends or their intention behind doing it was more so just like trying to create something fun. And so I urge you, if you are creating, you know, AI generated stuff is to just consider your intention behind why you're sharing it. For myself, one thought process I have when I'm sharing stuff is, like, I don't wanna share too much AI stuff because I don't want people thinking that I'm leaning so much into it that, like, it's a replacement for me sharing something authentically, and authenticity is a really important value for me. So if I think creating and sharing AI generated content is going to decrease the amount of trust and authenticity in what I'm putting out into the world, then I'm gonna probably put out a little bit less of that. Or make sure to, like, also supplement in real things. Or when I'm sharing that thing, like, how can I add more of a human element into it? For example, I have a post where I share some things from my desk, and I use ChatCha b t to make that photo. But, like, all of the things are pretty curated. It's not just your run of the mill desk setup. Like, I intentionally went behind each object, and it took me thousands of dollars and years to, like, arrive on the things that I chose. And I want to share that with people. And and I could just, like, take a picture and then talk about each item, but that doesn't really feel inspiring to me, and it isn't really the medium I wanna use to communicate that idea. So I have this, like, AI generated comic, and I can make it more personal by just in the caption talking about what each item means to me. And there's so much, like, tiny nuance that people pick up on. And I pick up on it every day. Like, I don't know how many times a day I see stuff that's like, it's not just about what's on your desk that matters. It's the relationship with the things on your desk. And it seems like ChatChippy Tea always does stuff like that. And I don't use AI for that stuff. Like, if if I wanna tell you something, I'm gonna tell you. Because AI is not gonna tell you better than I could tell you. But maybe when it comes to drawing, which I suck at drawing, so maybe it could draw the thing better for you, but I will explain it better through the words. And that's really important to me. So just something to be conscious of. Don't be afraid. Lean in. It's wild times we're living in. And if you have interesting use cases you wanna share about this new OpenAI image generation update, head over to thegreaterclub.com and share them with the community because we'd all be interested in seeing them. Hey. Let's talk about podcasting for a minute. While I mostly run solo episodes these days, it's just what works best for my show's format. One of my favorite things is actually being a guest on other people's podcasts and here's why. There's something special about having a direct focus and real conversation about the things I'm deeply passionate about, Whether it's the ins and outs of running my content agency or the creator economy or the entrepreneurial journey. And here's what matters most to me when I'm a guest, it's not about promoting myself or my stuff, it's about providing genuine value to your audience. So maybe your listeners are struggling with creating content, maybe they're trying to scale their business, or maybe they're just looking for insights about what it takes to show up consistently on social media. Whatever it is, I'm all about tailoring the conversation to what your audience needs most. I've been in the trenches. I've produced thousands of podcasts and social media videos. I've been running content clips for a few years now. I've been helping creators break through their blocks. There's a ton we could talk about. If you think your audience would benefit from a genuine conversation about any of these topics, let's talk. Just shoot me an email at hi@dylanschmidt.com. Thanks again for joining me today. Until next time, keep creating.