cross-posted from: https://beehaw.org/post/28095114
Vapor World: Over The Mind is a Soulslike Metroidvania with an evocative, painterly art style reminiscent of Lies of P and the narrative platformers of Playdead. According to Korean studio Alive Inc, the game had 100,000 wishlists just before its early access launch on August 19, mostly on the strength of a demo available since January 2025.
But what should have been a triumphant early access launch went pear shaped very quickly. At launch, few talked about the quite distinctive worldbuilding or the parry-heavy combat. Instead, all focus was trained entirely on something absent from the popular demo: very obvious AI use.
The problem is that there is no way of knowing (at first sight) if there was an earnest attempt. AI has a certain look to it, and that look has been associated with slop and shortcuts because of unbridled use of an imperfect technology. It makes sense that it would illicit a sense of revulsion. Also, people who lack the skills or time necessary to create a certain type of art are more likely to lack the ability or patience to properly judge the quality of that art. “Good enough” is rarely good enough.
Man there’s no way to know that someone made an Earnest attempt with traditional art either they could have just been tracing, stealing or repackaging.
If you require proof of every little thing up front before you judge it, then you don’t get to enjoy anything.
At some point you have to trust people at their word or you just don’t get to have fun.
That’s just unfortunately how society the real world and reality works.
I think this is a dangerous misconception.
I could feed locally run Minimax H3 a set of reference art, like the painted stills from this game, and it would exactly match the style. I could use start/end frames, motion masks, pose control, character sheets, voice and music samples, reference videos; I can get it to do whatever I want with extreme precision, and as long as I checked the output it would be indistinguishable from CGI.
The point I’m making is that a lot of people percieve diffusion as mashing a Stable Diffusion keyword prompt into a text box, and getting a sloppy result. Maybe thats still how shit corporate services like ChatGPT work.
But a LOT has changed over two years.
Theres no way people can definitively detect generative output anymore. We can IF its done lazily, but its quite possibly indistinguishable now, too.
You’re asking them to learn and understand rather than parrot… god speed.