ChatGPT Images 2.5 Streamlines Business Marketing

Jamie (0:00)

Welcome back to the podcast. Today, we have an exciting episode. Just yesterday, ChatGPT released Images 2.5, and they already have a pretty impressive image generator. I would say both me and Jayden use this one as our main image generator, if we need to make an image. But they just improved even more.

They have something called Sketch, lots of really exciting things to talk about today, and how you could actually use this to make money. Before we get into that, Jayden, can you tell me about our school community?

Jayden (0:31)

Yeah, every week, we have completely revamped our entire school community, and every week, we've been putting videos out for a couple of years. But now we have created an entirely new experience with way more content, and we have guides for every single lesson that we teach. We basically have it set up, so now you can go in and pick a different path that you're going to take, a different side hustle, whether that's growing and scaling the business you have, whether that's building a SaaS for the first time, whether that's becoming an AI consultant, whether that's doing automated AI music or podcasting. So there's a bunch of different things we've talked about in the past, but we reorganized everything and rebuilt it and are relaunching the school community's courses with updated stuff, updated information, so you're getting the freshest stuff there. And we truly believe this is something that will help you grow and scale your business now. So we're really excited about this. We have over 1,000 members in the school community. We've grown a ton over the last couple of months, up from about 300 to now over 1,300 members. So yeah, we'd love for you to join the school community. It's $19 a month, which is an incredible value for what you're getting, over 100 videos that Jamie and I have recorded on growing and scaling a business using AI tools. So if you're not already a member of the school community, we would love to have you join, and hopefully we will see you there. But talking about what's going on with OpenAI, I love the image models out of OpenAI for basically all of the businesses I run. I'm using it. It's interesting that Claude, which was kind of my go-to AI model for so long, didn't have image generation inside of it, and I felt like OpenAI just crushed it with their last model. But I've been really impressed.

They keep coming up with better models. And this new model, this new 2.5, has a bunch of really cool features to it. They specifically have a whole feature set where you can improve old photos. You can sharpen old photos, and you can change them. But they're very true to what the old photo looked like. So if you have an old picture of yourself, you can change what you look like or what you're doing, and it sharpens it really well. Which if you've used the, I don't know if you've done this with Canva, Canva has a quote-unquote AI upscaler, and it is flaming garbage. So I highly recommend using OpenAI for this, but the new model is pretty cool.

They have a bunch of demos that are pretty interesting where they're basically showing what the new Image 2.5 does with the same prompt versus what their Image 2 did. And honestly, I thought their Image 2 was great, but 2.5 really takes it to the next level with a lot more, just the images are higher quality, the scenes look better, everything looks a little bit more realistic. So it's kind of hard because with art, there's a fine line. But yeah, I think so far, they've done a great job with making some of the updates on this overall. They had one demo where they were showing Image GPT Image 2 versus GPT Image 5 And they said generate high quality, 150 high quality frames of a cube spinning. And they showed the Image 2, and it's kind of glitchy when it bounces around, and they are showing 2.5, and it looks super smooth. So just being able to really understand. I mean, that's just showing that from one image to the next, doing tiny frames.

It's basically they've created a video machine with their image machine because of how smooth all of this is going. But I don't know, what are you thinking about this, Jamie? How are you going to be using this in your businesses?

Jamie (4:13)

Well, I mean, one of the main major things that we use for here on the podcast is actually thumbnails for YouTube and for any place there would need to be an image thumbnail. And of course, Jaden and I, in the past, in these old thumbnails, don't really look that much like us. It kind of has our likeness, but they're kind of these own AI characters that is generated. But now with the new model, similar to how it showed the cube spinning, it has a full understanding of our likeness and actually keep it that way. So then our character looks consistent across thumbnails. This could be also really useful if you are doing graphic design or advertising, and you need to come up with a bunch of different creatives for an ad. Having, keeping a consistent character or consistent look of the same person across would be very important. But I think it's really cool. A lot of the demo images are showing things that really just look creative and actually well designed.