News Valve Allegedly Bans Games on Steam From Using AI Art

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
I can totally see EPIC now saying "we welcome all AI generated art" and just diverting all the money from their, ehem, foreign investors into filling the lawsuit holes so they get more publishers in their platform xD

Let's see if time says I'm wrong or not 😀

Regards.
 
AI was a tool built to mimic human ideas and concepts.
Like any other tool humans have built it can be used for good or evil with many use cases in between.
To say AI was plagiarizing would be like accusing a hammer of being good at smashing your finger.
The hammer was designed with the express purpose of smashing your finger!
 
I don't see a problem here with AI - at all!

1. Commercial use of copyrighted material without paying licencing fees has always been prohibited - with or without AI - so nothing new here.

3. Of course, one could prompt the AI to generate pictures / assets that would sufficiently differ from copyrighted material in order not to infringe any copyright laws. And this requires not more than proper programming/"prompting" the AI. Just as a developer studio would have to instruct ("prompt") a contracted human graphic designer to create visuals that would not have copyright-lawsuit-suitable looks.

The AI can train on any material - copyrighted or not - as long as its output has been properly shaped and is sufficiently different from the copyrighted training materials.
 
  • Like
Reactions: BX4096
It's not that at all, it's not intelligent or sentient or anything thing remotely close to that. It merely has created a massive data model that links similar data to each other.
The analogy is apt and very much by design. Today's AI falls under a category of neuromorphic computing, where the techniques being employed are modeled on key aspects of animal brains and how they learn. We humans like to believe intelligence is a lot more complicated and mystical than it really is.

ChatGPTs for example is a 50 thousand element 12 thousand dimension matrix that contains every word in the English language and it's relationship with every other word in that language. By analyzing input text it can sequence the next most likely word and so forth until it gets an output that is then sent out as text.
That's how it's trained, but doesn't adequately describe what it's doing. It's not simply a giant cross-correlation matrix between words. It's actually capable of modeling higher-order concepts, or else there's no way it could hold as much knowledge or perform such complex tasks. Did you know it can even do arithmetic and simple reasoning? You don't get that capability from a simple and direct correlation of words.

All SD and other art tools is doing is analyzing the input words, determine what effects, filters or other material those words most likely mean, then fetching them and performing that operation.
That's not consistent with any papers I've read on GANNs or image generators. I trust you have a good source on that?

It still has to fetch source material,
There's no "fetching". The knowledge of what certain things look like is already baked into the model.

and if the creator does not have the legal rights to that source material, then a copyright violation can take place.
A copyright violation isn't an abstract thing. You have to actually demonstrate that it copied specific likeness from specific copyrighted artwork. Given how generative AI works, that threshold is unlikely to be met.

Now there are rules about fair use and derivative works, which might eventually encompass AI assisted material, but they aren't there yet.
The claim is that the model is a derivative work. I don't expect that bar will be met.

Kyle Hill did a very good breakdown in laymen's terms on what is really happening.
Based on what you've claimed so far, I think you should seek a better source.

If you have no functional expertise in a field, how are you qualified to judge how well someone does at explaining it? You can't know whether your understanding is correct without testing it in some way.
 
Last edited:
  • Like
Reactions: ottonis
If you post something online like this comment, IT'S PUBLIC. How on earth do people not know this is mind baffling to me ........
Putting something on the internet does not necessarily make it public domain. Even your example of posts to this forum is probably wrong. I haven't checked, but I assume the website's publisher assumes ownership of everything we post on here.

The art industry won't have an impact, this will sieve the pool and leave behind people who actually/genuinely love art rather than seeing it as a cash cow.
People said the same sorts of things, back when MP3 piracy was rampant. The problem is that a musician can only hone their craft to a certain degree in their spare time, and most can't make a living off of performing. There's a sizable set of artists who depend on publishing/streaming royalties in order to be able to make music their full time gig.

(Real) Artists will be fine.
I'm sure the story with visual arts is similar to the case of musicians. It doesn't mean we won't have art - just that artists will become an endangered profession and mediocre and "uncanny valley" art will be a lot more common.

For example, I would much rather get an oil painting by a real human rather than a mechanical machine into the future.
Real oil paintings will (eventually) get more expensive and the typical quality will be lower.

Anime Unity Template garbage is being spammed onto Steam, yet A.I generated textures is not allowed?

Should we even respect Steam as a knowledged platform? Seems more like a decade long terrible student software project, that steals 30% off of developers, which is more than some actually criminals charge....
You're confusing the issue - this is about rights, not quality. Those templates are licensed assets. What Valve is clearly concerned about is their exposure to copyright infringement lawsuits.
 
Last edited:
AI was a tool built to mimic human ideas and concepts.
Like any other tool humans have built it can be used for good or evil with many use cases in between.
To say AI was plagiarizing would be like accusing a hammer of being good at smashing your finger.
So, they're not accusing the AI of plagiarism. People making accusations (which is a slightly different than this article) are either accusing the people who built the AI model of misappropriating copyright works or of otherwise using them to build a plagarism machine.

It's hard to reason this out by analogy.

one could prompt the AI to generate pictures / assets that would sufficiently differ from copyrighted material in order not to infringe any copyright laws. And this requires not more than proper programming/"prompting" the AI.
That's where I think the future lies, due to some of these specific concerns.

It's not just a matter of what prompts you give it, though. You need to actually design & train a model to be able to devise different styles, compositions, or subjects than what its seen. Many image generators already seem to have a pretty good handle on immutable rules vs. what elements and aspects can vary. So, it kind of "knows" which knobs to twiddle, and just needs to come up with combinations of parameters that it thinks are fairly novel.
 
Last edited:
  • Like
Reactions: tamalero
You're confusing "artificial intelligence" with real intelligence. AI merely refers a system that seems to behave intelligently. In the 70+ years the AI field existed, they never set such a threshold for themselves as the one you're constructing.
(AI) The subfield of computer
science concerned with the concepts and methods of symbolic
inference by computer and symbolic knowledge representation
for use in making inferences. AI can be seen as an attempt to
model aspects of human thought on computers. It is also
sometimes defined as trying to solve by computer any problem
that a human can solve faster. The term was coined by
Stanford Professor John McCarthy, a leading AI researcher.

The famous Turing Test established the first gold standard for AI: that it merely be good enough to fool a human into thinking they were chatting with another person.



Creativity isn't anything too mystical. It's not hard to approximate by running some random noise through enough filters to reject the rubbish. At some level, that's even what your brain is doing.
Agree, the current "AI" is just a knee jerk name for a feature. Aka a marketing name.
Current AI is no real AI.

Reminds me of RTX of Nvidia.
It is still not fully raytraced. And its a marketing name.
 
There is nothing wrong with A.I generated art if it has been trained on public data. If you post something online like this comment, IT'S PUBLIC. How on earth do people not know this is mind baffling to me ........
I'm sorry but this take is one of the silliest things I've ever read.

Just because its ONLINE, does not means you have the right to REHUSE or MODIFY.
The fact that is is online is because you can CONSUME AS IS. Aka see/read/listen and enjoy.
 
  • Like
Reactions: bit_user
Agree, the current "AI" is just a knee jerk name for a feature. Aka a marketing name.
Current AI is no real AI.
Define "real AI". With authoritative sources - not just a definition you pulled out of thin air.

It's funny to me how everyone suddenly seems to feel like such an expert on AI. Yes, you have an intelligent brain. Does that mean you know anything about how it works? I drive a car - it doesn't mean I know the first thing about designing or even fixing them.
 
  • Like
Reactions: TJ Hooker
Define "real AI". With authoritative sources - not just a definition you pulled out of thin air.

It's funny to me how everyone suddenly seems to feel like such an expert on AI. Yes, you have an intelligent brain. Does that mean you know anything about how it works? I drive a car - it doesn't mean I know the first thing about designing or even fixing them.
The plain definition of AI and its name is already what defines what AI is.

Current "AIs" aren't aware.
They do not get us, the just process based on specified view points, keywords, etc..

ID say, current methods is just brute-forcing massive amounts of data.
 
The plain definition of AI and its name is already what defines what AI is.

Current "AIs" aren't aware.
They do not get us,
See, that's the problem. People latch onto the word "intelligence" and reject anything that seems in any way inferior to a human, even if it can far outpace humans in other areas. What they're thinking of is referred to as "Artificial General Intelligence".

The field of AI is not principally concerned with trying to clone human intelligence. One way to think about it is simply looking at approaches to problems which defy straight-forward conventional algorithmic solutions.

Consider this:

Think of a wolf. They're social, they can strategize, work in teams, figure out puzzles, etc.

How about crows? They can observe and adapt to human behavior, recognize specific people, solve problems, etc.

What about octopuses?

Are these animals intelligent? I would say so. There are lots of examples of intelligence, in the animal kingdom. Even some social insects, like bees, who can argue with each other, make collective decisions, and problem-solve.

I would say intelligence is a more fluid concept than we might like to think. That makes it difficult to establish a clear, bright line between intelligent vs. not intelligent.

the just process based on specified view points, keywords, etc..
That's not accurate. You'd do well to educate yourself more on how LLMs really work. If you're the sort of person who likes to understand how things work, you might even find it fascinating.

ID say, current methods is just brute-forcing massive amounts of data.
We've had lots of data, before. What we didn't have was the computational capability or the algorithmic techniques to build systems which model higher-order concepts. That's one of the truly remarkable things about LLMs - their prowess at knowledge-representation.

They don't only operate at a word-level, but model and operate on abstract concepts. It's even more fascinating that they learn these concepts without anyone having to explicitly specify them. It seems to be a natural byproduct of seeking representational efficiency, which is probably similar to what happens in human brains.
 
Last edited:
The guy making the claims discussed in the article didn't seem to provide any examples of what sort of AI-generated images Valve took issue with, aside from describing them as "a few assets that were fairly obviously AI generated". For all we know, they may have generated images that looked like blatant ripoffs of copyrighted characters or redrawn art from existing games.

And what kind of games does this guy even develop? They were careful to avoid providing any clues about what the game was, going so far as to block out the game's name from the screenshot they posted of Valve's message to them. It might just be a repackaged asset flip with some AI images inserted to hide the fact that they didn't put any effort into developing an actual game. According to them...

That sounds a lot like shovelware to me, if their entire trailer consisted of AI-generated images.
It seems I was right. What all the news sources regurgitating this article didn't look into was that the "developer" is apparently someone who was attempting to churn out shovelware for a quick buck. I happened to open their Reddit feed again from my history when looking for something else, and their latest post states...

I created an AI art generated Hentai game, which is just simple puzzle game with Hentai images. Took me like a week of learning basic game dev, a week of coding, and a week of making the art, and I had a game up on Steam making me 100s of bucks.

Have a video outlining my process:
That post was made as a response to a thread titled "ULPT Request: How can I legally profit off stupid people?" In the video he linked, he actually refers to the type of game he is making as "low-effort cash-grabs" and brags about how he already made a thousand dollars off one being sold for 74 cents a copy. So that's the real story here. Valve didn't want another of his questionable shovelware slider puzzles on Steam that was identical to his previous title, just with different AI-generated images slotted in.

Also, the rejected game had a title that is extremely similar to the name of a popular anime that recently had a game release a few months back, likely hoping that his shovelware would come up in the search results of people looking for that game.

I think that should tell you everything you need to know about why Steam rejected his game.
 
  • Like
Reactions: bit_user