https://arstechnica.com/information...eometric-scenes-mesmerize-social-media-users/
This recent article goes over the technique to generate these kinds of images. Maybe this will change your mind about how truly accessible it is for any one of us to create images like the one Marky shared. And by the way, I'm not putting the image down or anything but I don't think you fully understand how powerful this "brush and some paint" truly is. There's a lot less imagining than you think that goes into making an image like this. The imagination you're speaking of is simply being curious enough to try these tools out; I feel like you're taking my comments as some kind of attack on art or something.
Up close, the image may not appear out of the ordinary (although you do see some typical AI image generation artifacts like a one-legged man). For me, the hidden image becomes much clearer after backing away from the image.
So don't get me wrong, this technology is impressive but it truly is no exaggeration that a 5 year old can probably do this. I mean, they'd have to be pretty smart but the more salient point IMO is that a person with no artistic talent (i.e., me) can generate images that sometimes turn out pretty decent.
Some people will debate that there is a lot of "work" that goes into generating a good image through Stable Diffusion or MidJourney, but I don't think it's that difficult. This is exactly why we're seeing so much of this AI-generated content (even for mainstream commercial projects).
View attachment 16151
This took me 30 seconds to make (aka write a prompt of a few words), and another 30 seconds for my PC to process my prompt and generate this image.
To bring this all back on topic...
View attachment 16152
This took me around 6-7 minutes to generate (had to go through a few iterations). I'm pretty happy with how this turned out. Again, it's not perfect and sometimes you maybe end up with 4 fingers instead of 5, hence the need for slight tweaks generally. As I said earlier, I have zero artistic talent. I see the human user simply as a guide when using AI image synthesis tools.
Actually, I surprised myself with how well this came out and I might do this for some of the players eventually.