I partially disagree, but the content itself absolutely is what attracts people in the first place and gets them to click. However, in terms of the people who choose to more closely follow an artist or support them, or praise them for their work, the human element absolutely plays a part. You can’t follow or support everyone, so you’re more likely to prioritise those who you feel a more personal connection with and/or have the deepest appreciation for.
I’m pretty sure AI generated art was initially meant as a tool to assist artists in creating their own works either through posing, ideas for scenes, color/lighting effects, etc. and not to outright make a piece entirely out of AI and post it as is or with only minor edits to what is technically not actually art but a generated mash of data.
I feel like this a bit of a naïve viewpoint given the implementation of it. If AI generated art was made with artists in mind, they would have asked for permission for stuff or only used public domain images for training, and it would look to supplement the artistic process or help an artist learn.
The idea would be to allow artists looking for idea to browse through it, get their ideas, and create their own artwork out of it. It would keep it separate from the actual works on the site so as not to flood anyone’s searching with them and would allow people to gather free use materials to use as they see fit, within reason and without infringing upon any actual artists who’s original works might have been used in those generations.
When an artist uses them the AI assisted tag would still apply and then they can link back to the source(s) that they used with something like a >>AI####… system like how we link to normal images with the normal numbering system. This would allow strictly AI generating users to upload more freely (perhaps still with some limits to prevent garbage upload bombs) and give them a place to discuss prompts and such amongst their selves while keeping those who don’t like AI stuff to simply not visit that section of the site.
If it’s purely about inspiration then AI art does very little IMO to help with this, at least with the way you’ve suggested here. I’m not saying that you can’t be inspired by an AI image that is good enough to be appreciated, but inspiration is literally all around us in many forms. You can be inspired by things you see in real life, or on the Internet, but I’m not sure if I see the benefit of having a catalogue of AI images to choose from unless I’m misunderstanding the point.
Secondly… in regards to how some people view ratings solely as people liking a thing in it’s entirety regardless of content, I would like to suggest a system where you actually have to view an image to rate it/fave it. Not simply seeing the thumbnail in your search results but actually opening it and seeing the full thing to properly know if you like it or if it’s just the small box of search result colors that happens to appeal to you. This would allow people to actually see if an image is what they are making out in the thumbnail or if it’s just what they want to see and are simply imagining it as something they like.
Great idea in theory but I’m not sure on how it would pan out in reality. Many people on the Internet are unfortunately inherently lazy and would continue to judge art based on how it looks at a thumbnail size anyway. The only difference is that they’ll click on the image, immediately click upvote/downvote/fave and then leave… if they can even be bothered to leave a rating anymore.
This is what happens when you focus on having more content rather than good content - people become jaded and are less inclined to spend time to properly look at something unless you manage to grab their full uncontested attention. And yet, they’ll say they want more content when there’s already loads of things out there, more than they could reasonably go through and digest.
