Having been into generating my own AI images since March of last year, I can say that I am enjoying my experience, creating things that you would not normally encounter in every day life, for example, a man hiking nude in the Swiss Alps, or a half-man, half-amphibian hybrid creature. It's just interesting to see how it might turn out. And if you don't like what the AI creates, then all you have to do is click the 'generate' button again to see what other iterations of your image the AI will create.
It doesn't scare me, but when I first started, I couldn't believe how real looking some of the images were, especially faces. I had to keep reminding myself that the person isn't real, the situation isn't real, nothing is real about these, but because of AI training and people publishing AI models that seemingly create more realistic images, it's getting harder to tell what is real and what is not.
If anything, I have a deepened appreciation on how to describe something. For example, instead of saying "create an image of a man in a blue suit with green pockets", you could tell the AI to "create an image of a man wearing a blue pin-striped suit, a sophisticated and tailored ensemble, exuding a classic yet contemporary charm. The fine, navy blue fabric is adorned with subtle, vertical pinstripes that add a touch of elegance and formality to the overall aesthetic. The suit jacket features a well-defined silhouette with notched lapels, providing a polished and professional look. What sets this suit apart is the unexpected and stylish twist of vibrant green pockets, seamlessly integrated into the jacket. These green accents add a playful and unique element to the ensemble, creating a subtle pop of color that catches the eye without compromising the suit's overall refined appearance. This combination of traditional tailoring and a bold color choice makes the blue pin-striped suit with green pockets a distinctive and fashion-forward choice for those seeking a modern edge in formal wear", and see what the AI comes up with. You *might* get similar images from both prompts but likely to get better results from your more descriptive request. (Curious about how something like that would look...I created an image... it didn't create the green pockets, because the AI didn't understand where to put them, but it did understand to create something green in the image. You get the idea...)
![00069-20240125173335.png]( "00069-20240125173335.png")
11 months ago I never thought I'd have this many AI generated images on my site and gotten to a point where I think I'm getting pretty good at describing what I want to see and the outcome when the image is created. I'm enjoying myself, and I hope what I create and post on LPSG that some of you like what you see too, so I'll likely continue creating them.
