Also, for those of you curious how to do it. I'm not sure what others are using, but I use the "Bing App." It has a "CP" button (Co-pilot) at the bottom
. Click that and its opens a chat screen, it looks like messenger or a regular text message conversation
.
Make sure GPT-4 is turned on. Then just ask it to draw something simple. . . (please draw a man made of metal wearing Tennessee Vols Colors)
It'll make the image. See what you want to add, then just ask it to add it. (Please make him stand in front of Neyland Stadium. . .or make him a big-game hunter)
Look at the next image it makes and see what you want to add and just ask it to. (Make him stand in a room with exotic animal heads mounted on the wall. . . or put a tsunami in the background, etc)
Here's what's weird. . . you seem to get much much better results if you're nice to it, saying please & thank you, talking to it like a normal person would and telling it "great job" or "you're very impressive." Sounds weird, but I get much better results and it will let me alter the picture many more times when I say stuff like that. Sometimes it'll create 12-16 images for each prompt, instead of just 4, when you compliment it. It never did that before when I just gave it orders. So strange.