3 ms·
I use AI art for something that would never be worth the time for a human to do it: solution images for a daily word game that tie the daily solution words toge
by furyofantares 4y ago
I use AI art for something that would never be worth the time for a human to do it: solution images for a daily word game that tie the daily solution words together. It's a popular feature that takes me about 5-10 minutes per day to produce a good image.
So that's new demand for art created by it being so accessible. Most of the art has some very obvious flaws and deficits; much of which could be cleaned up quickly by a human. I certainly imagine new jobs for artists that use this tool to assist in very quick jobs, for applications higher value than my use case but which still wouldn't exist if a human had to do all of the work.
- UberFly 4y agoThough you can easily see a quick jump to a point where the creating and then the tweaking doesn't need a human involved.
- furyofantares 4y agoI can't easily see that, no. That's a very presumptive way to word your speculation. I don't have much of a prediction except that it seems a long long ways from there not being value a human can by working on prompts and cleaning up images. For sure there's already applications where you don't need a human involved per image. But for applications where it makes sense to spend 5, 10, 60 minutes of a human's time to improve the output, I think the baseline output could improve drastically and as long as humans can still make improvements within a timeframe that makes sense for the application, it will still make sense to do so.