> do you think there's something lost by eliminating scarcity? Certainly whatever these models can output just from text plus a bit of work/skill on the prompt

> do you think there's something lost by eliminating scarcity? Certainly whatever these models can output just from text plus a bit of work/skill on the prompt plus a bit of selection will not be very valuable. But it will see applications that don't need something very valuable, and where otherwise a piece of art would be not be affordable. I run a daily word puzzle game, and every daily puzzle comes with an image that helps tie the theme together. This is a low value piece of art that's nearly free for me to make, which wouldn't exist otherwise. But the other thing is that people will make scarce and valuable things using generated art. Right now we're seeing tons of art where the direct output of a model is posted after less than an hour of work on prompts and selection by a person. But when more people start combining all of these tools, spending dozens of hours on an image that's made out of components that are generated using AI tools, and other components that are manually edited or created, I think we're going to see a lot of brand new art and brand new skills. There's always going to be people who pour their creativity, skills, and time into something. I can't wait to see what people can make when they do that while incorporating all these new tools that are being created into their process.