Avatar

Bruh. (Gaming)

by stabbim @, Des Moines, IA, USA, Thursday, March 26, 2026, 23:37 (4 hours, 13 minutes ago) @ Kermit

So is the AI issue the ethics around that. I think anything AI has become a hot button, and like many other issues, people act like there is no middle ground.

When it comes to generative AI, unless someone can prove to me that their model was trained ENTIRELY on content whose creators opted in to such use, then I don't see how there can be a middle ground. It's theft that the law mostly hasn't caught up to yet because the people making the law are either too old to understand what's happening, or they've got financial interests in allowing it to continue.

On top of that, it's theft that uses enormous resources and generates soulless "content" that's inevitably going to be a middle-ground amalgam of things the model has previously seen, but gets passed off as "art" by people who have no appreciation for creativity and think art is a product that you purchase by the hundredweight. That's why people don't like DLSS 5, by the way - it's got nothing to do with how realistic the graphics do or don't look, and everything to do with it not being what the artists, actual humans with ideas and points of view and things they wanted to communicate, worked hard to design in a specific way (the on/off comparison images are humor, and they signal to others that you're on that side of the debate, but graphics quality was never the actual point), only for some clueless tech evangelist to stumble drunkenly into the studio, spill a bucket of paint on the canvas, and declare that it looks better that way before hiccuping and passing out.

You will sometimes hear an argument that it's "just like a person learning," by the way. First of all, no it isn't - while I will grant you that what a person learns informs their future cognition in a big way, it is possible, however rare, for us to come up with new applications and occasionally entirely new ideas. The old nothing new under the sun argument is a useful lens to look at the world through and understand that you're always building on what came before, but it's not entirely true. Sometimes people do have thoughts that haven't existed before. These neural networks can do no such thing. Second of all, none of us humans were built by a corporation for the express purpose of ripping off ideas as a business model. We exist, and we learn, and perhaps we even plagiarize things sometimes! But we are not a commercial product which was designed, invested in, built (and shoved into people's lives who didn't ask for it), solely for that purpose. And what exactly does all this theft, and sapping lazy people of the creative experience, accomplish? A bunch of tech bros get to drive another investment bubble. Cool. Cool cool cool cool cool.

* There are other forms of machine learning which can be ok ethically. I'm looking forward to developments around medical diagnostics, protein folding, all kinds of stuff, and I wish more of the investment money was going in that direction. But those things aren't usually what's being debated. I only mention them to be clear on where I draw the line and why generative was in italics.


Complete thread:

 RSS Feed of thread