The binary thinking around "uses AI" or doesn't is going to burn down indie gaming if it continues. Review-bombing games for having 0.1% AI-generated content doesn't benefit anyone but it's the kind of pointless crusade that gamers love.
The issue is similar to adult games. There are some very high quality stuff, but most of what you'll find is low effort cheap releases. Meanwhile, the good stuff takes time to leverage and scale.
Steam allowed AI games (practically speaking) less than 2 months ago. Unless an indie was sitting and waiting for Valve to change its mind, there's no way any quality game would come out in that time, AI or not.
Usually a lot of effort is put into building and tuning procedural generators. They are often game specific. Using a generic AI to generate assets and replace artists is a bit different.
If the results are still creative and compelling, what does it matter what the method used to get them is?
If people use genAI and produce boring, uninspiring content, their games won’t succeed. Just like if they use hand tuned procgen algorithms to make boring uninspired content - or if they manually create boring uninspired content for that matter.
If they can make compelling content using genAI or rolling their own procGen or hand authoring everything then great! Compelling content is the goal. People don’t play effort they play results.
I'm amazed that anyone thinks this is surprising. The biggest differentiator in terms of games' perceived quality is the quantity and overall 'slickness' of their content - artwork, characters, backgrounds, copy.... Of course developers are going to take the shortcut to add more high quality content to their games.
(Author) Good point! I picked the top categories by mentions. 24 entries that mentioned "code" or "copilot" (out of all total disclosures), and a third of them actually went out of their way to state that there was NO AI code gen; typically like so:
> it is not used in the game itself in any area: 3D models, code...
I suspect that a more rigorous perusal of the metadata (i.e., more than those quick search terms) would turn up some more, but either way, it seemed like such a tiny fraction of the whole.
Well shit, all of my code from the last 3+ years would need a trigger warning then :D
I've found Copilot (and its like) to be essential in the way I work.
It's a lot faster to ask an AI assistant to do the boring repetitive bits + me glancing through them than me writing and checking documentation and writing and getting bored and my ADD kicking in and now I'm on Wikipedia reading about some weird castle a baron built on top of a mountain just because. =)
Glad to see Valve figured out a fair solution to this. Gen AI opens up game dev to more people which is a good thing. Coders can use it silently in the background with impunity but visual and audio creators cannot. I expect as the tools become more widespread we will see more games and a lot of them will not be good, because the gameplay and the current trends the gaming community, determines what is successful .
The binary thinking around "uses AI" or doesn't is going to burn down indie gaming if it continues. Review-bombing games for having 0.1% AI-generated content doesn't benefit anyone but it's the kind of pointless crusade that gamers love.
The issue is similar to adult games. There are some very high quality stuff, but most of what you'll find is low effort cheap releases. Meanwhile, the good stuff takes time to leverage and scale.
Steam allowed AI games (practically speaking) less than 2 months ago. Unless an indie was sitting and waiting for Valve to change its mind, there's no way any quality game would come out in that time, AI or not.
Steam only banned AI in June and I'm aware of one game released before the ban that got roasted.
Funny, generative maps used to be praised for their replayability or uniqueness. But now that it isn't simply algorithmic it is worth an uproar?
Usually a lot of effort is put into building and tuning procedural generators. They are often game specific. Using a generic AI to generate assets and replace artists is a bit different.
If the results are still creative and compelling, what does it matter what the method used to get them is?
If people use genAI and produce boring, uninspiring content, their games won’t succeed. Just like if they use hand tuned procgen algorithms to make boring uninspired content - or if they manually create boring uninspired content for that matter.
If they can make compelling content using genAI or rolling their own procGen or hand authoring everything then great! Compelling content is the goal. People don’t play effort they play results.
There's a whole lot of people very upset that the prescribed order jobs would be replaced by automation isn't going as they assumed.
I'm amazed that anyone thinks this is surprising. The biggest differentiator in terms of games' perceived quality is the quantity and overall 'slickness' of their content - artwork, characters, backgrounds, copy.... Of course developers are going to take the shortcut to add more high quality content to their games.
I would assume real number is much higher.
Games have for decades used generation as source for content, now available methods are just more complex.
Well it's just the amount that declare it on the partner portal and their store pages and are currently publicly viewable.
Said number is roughly 1354 currently if you go around counting things like DLC and Demos distinctively.[1]
The majority being games with ai-generated assets, with a small number (~99 currently) being live-generated content like chatbot interactions.
https://steamdb.info/search/?a=app_keynames&type=-1&keyname=...
It's interesting that Valve's policy requires disclosure of AI generated code:
> Any kind of content (art/code/sound/etc) created with the help of AI tools during development.
But that none of the categories the author of this post identified included code, only visual, audio and text content.
(Author) Good point! I picked the top categories by mentions. 24 entries that mentioned "code" or "copilot" (out of all total disclosures), and a third of them actually went out of their way to state that there was NO AI code gen; typically like so:
> it is not used in the game itself in any area: 3D models, code...
I suspect that a more rigorous perusal of the metadata (i.e., more than those quick search terms) would turn up some more, but either way, it seemed like such a tiny fraction of the whole.
Well shit, all of my code from the last 3+ years would need a trigger warning then :D
I've found Copilot (and its like) to be essential in the way I work.
It's a lot faster to ask an AI assistant to do the boring repetitive bits + me glancing through them than me writing and checking documentation and writing and getting bored and my ADD kicking in and now I'm on Wikipedia reading about some weird castle a baron built on top of a mountain just because. =)
Glad to see Valve figured out a fair solution to this. Gen AI opens up game dev to more people which is a good thing. Coders can use it silently in the background with impunity but visual and audio creators cannot. I expect as the tools become more widespread we will see more games and a lot of them will not be good, because the gameplay and the current trends the gaming community, determines what is successful .
GenAI will continue to be used. People will just learn to hide it better. Eventually it will all be GenAI and you’ll be none the wiser.