Video games made with AI (especially if it's the images) should be rejected.

General Discussion
, ,
65

Well, this is the first time I’ve been accused of writing using an LLM. Should I be flattered?

I suppose I should have expected this to happen eventually, considering that I’ve stubbornly refused to give up the use of the em-dash. I used to type it as a double hyphen (–), but in recent years, I discovered that I could access the single-character version by using Win+; to bring up the emoji/symbols menu, so I switched to doing that to save characters. Perhaps to my detriment.

However, regarding your accusation about quotation marks, let me invite you to try this: Use Ctrl+F on this page (hit it twice to get past the site’s built-in search function) and search for the quotation mark character in people’s replies.

Notice anything?

Yeah… they ALL look like that. Including here, in a message from sypern, who has always been strongly anti-AI (not just in this topic), and whose message was quite short and omitted the use of caps in some cases. Not exactly the kind of reply you’d expect from an LLM.

This is what their reply looks like on my screen, complete with the “weird” quotation marks:

image
image778×255 13.7 KB

So yeah. That’s fascinating. I didn’t know the forum did that. Learn something new every day.

As for me, I like to yap. Sometimes I get long-winded too. Sorry. If you need a sample of my writing from before the rise of LLMs, here are a couple somewhat-meaty posts I made on this site in bygone years.

EDIT: Oh, and would you look at that. The forum AUTOMATICALLY converted my attempt at a double-hyphen into a dash (maybe an en-dash? I’m not sure). Wonders never cease! Feel free to try it yourself.

This is my attempt at a double hyphen: –

And this is me using the Win+; menu to type an em-dash directly: —

EDIT2: Yep, looks like double hyphens get converted into en-dashes. Again, fascinating! But I’m afraid this is just how I write. Sorry.

2 Likes
66

Acting as “we need to learn to use it” already admits defeat. When people use it for one thing, then another thing is fine. If people use it for the images, who’s to say they didn’t ask it to write the script? Who’s to say they didn’t use it to generate the code?

In response to people being hated for using AI, people also were cheering for the “entitled” artists to lose their income because AI was better. Artists have long been harassed for “being too pricey” when artists are underselling themselves, again, due to how art isn’t seen as important. It’s seen as disposable. Art is free to see online, so the people who make it are worthless in these peoples’ eyes. With how much it has been pushed and shoehorned into everything, including things that 100% does not need generative AI, it’s out of control.

Ethical generative AI will not happen because of the people and companies who are funding the massive bulk of this stuff.

“There’s no ethical consumption under capitalism” is a phrase that gets thrown around a lot. It’s in reference to “don’t beat yourself up for not being perfect, try to limit your impact.” Generative AI uses WAY too many resources for how it’s not needed.

The assistance of an unthinking, unfeeling AI is worse than nothing. It’s worse. I’ve been worried about this since early 2023, and unfortunately a lot of my fears with the technology have been proven correct.

Humoring the idea that it’s a good thing is going to make things worse down the line, and the idea that actual human art will be replaced because art is seen as a disposable product and the generative AI machine can create passable works FAR faster than any human can will push human creativity out of what’s accepted. People will be ostracized and pushed out for not working fast enough. It’s scary. It’s scary and sad how willingly people are throwing this very important bit of humanity away.

5 Likes
67

A lot of people don’t realize that this sort of stuff cannot be given the benefit of the doubt for being used with good intentions. Think about why so much money has gone into it. Think about what the people and companies funding this are okay with sliding.

It’s dangerous to play devil’s advocates to people who will destroy peoples’ lives when they’re no longer profitable. Big companies like Microsoft, Google and Blackrock do N O T have ethical AI or fair treatment of human beings in mind when pumping billions of dollars into this.

4 Likes
68

But have you considered that the slop machine is convenient for me, personally, in the present moment and that I will cheerlead for anything that stops me from having to spend time thinking, freeing me up for dates with rosie palms?

Why use many thought when few do trick? /s

I think that enough reasons to avoid or limit use have been given so far to the point that minds have already been made up by most people on such tools. At the end of the day, we as a forum are never going to be the deciding factor in whatever fate befalls humanity.

There are ways to use such tools in non-destructive ways on a smaller scale, but I think anyone that doesn’t acknowledge the bigger picture is kidding themselves. As models get bigger, more complex, more resource-hungry, prices will go up. The tools handed out for free will be priced and gatekept like everything else. Power prices aren’t going lower any time soon.

And yeah, maybe the solution to having lots of poor old solo devs is for devs to work together more when they want to do something more ambitious? I would not be sad if there were half as many games, but each game that was made had twice as much content, or twice as much polishing. Or half as abandoned, I suppose.

3 Likes
69

There’s passionate and then there’s pedantic. Seeing just rebuttals in rapid fire posted one after another reads to me as not having a discourse and consuming the thoughts of others and just trying to strike them down. I don’t see the value in that vs. trying to actually read and understand each reply, consider their viewpoint, and further the discourse by trying to continue the conversation building on those thoughts, mixing them with your own, or having any admission that you may be able to see their viewpoint. It just comes off as an attack/vendetta to me.

In contrast your posts have been insightful and furthering the discussion, summarizing it, or reinterpreting it in some way.

I just wanted to reply to you as I appreciated your insights into the topic (but otherwise plan to checkout of this thread). I agree that there’s a lot of problems in the world that need a lot of people to get involved, but is indeed easier for most folks to do nothing which perpetuates the issues.

I do look to things like Star Trek in the future where Holodecks mystically create everything at their command, there is a lot of power there. But at the same time that’s not where we are now, and while in Star Trek they don’t bat an eye and they still value creativity and art. That’s not what’s been happening here either, and it’s capabilities are way overblown in many cases. Meanwhile real people are being effected in ways that are counterproductive to society (as many things have been as of late).

And as you said they’re trying to build an audience but it’s costing them money, none of this stuff is actually profitable for them. Like what practical purpose to the Sora 2 social network have??? It really seems like they’re trying to go down the ads route now which just means it’s all just smoke and mirrors or a flash-in-the-pan if that ends up how they’re trying to make their business profitable. But this is basically all of what Hank Green’s video stated so well:

70

Honestly? Whether they admit to partial AI use or not, you can’t know for sure that they didn’t use it. I mean, look at what just happened to me: I literally just got accused of using AI to write these posts! Granted, part of the accusation was based on an incorrect assumption about how the forum displays certain characters, and I was able to provide writing samples to show that I really do write like this (also evidence that I don’t need AI’s help to write), but still. Someone apparently had trouble telling my writing apart from that of an LLM (which is perhaps a bit depressing). Can you be 100% certain that I didn’t write any of my posts using AI? Can I be 100% certain that YOU didn’t? (I believe you didn’t, of course, but you get my point.)

You’ve identified plenty of problems with AI already—and I agree with you on many of them. (And apparently, now I have to add AI ripping off my writing style to that list of concerns.) I know that the big picture is scary, but I can’t do much about things on that level. All that I really have a say in is how we behave in communities like this. And I don’t want our community to be at each other’s throats over this. (Though these frequent AI discussions always seem to have that effect…)

We’ve already covered how AI can be helpful for solo devs. You don’t have to give up your overall anti-AI stance in order to acknowledge that. Would it be better if they could do everything without AI, perhaps by banding together, as @cheddar suggests? Probably. But just because they did use AI doesn’t mean that the parts they worked on themselves don’t have value, or that they didn’t learn something in the process of making a game, even if they only made part of it.

And call me a fool if you will, but I really did mean what I said about human creativity. You say that you’re afraid people will lose part of their humanity—their artistic side. I say, because it’s part of their humanity, they never will lose it, even if they try. That doesn’t mean the AI apocalypse won’t happen, but for as long as we survive, we’ll always be creative.

Now if you’ll excuse me, I’m going to go have an identity crisis over my writing style. :melting_face:

2 Likes
73

dude, that shit was trained on humans, of course some actual people type like that, if they didn’t the ai slop machine wouldn’t have anything to pull from

3 Likes
75

Make a text adventure. Exchange money for goods and services. Make an interesting game with bad graphics so an artist DMs you and says “need some help?”

The idea of the solo dev is a fantasy. Collaborate with other people.

5 Likes
76

The reason why AI discussions end in people becoming overly hostile is because the technology has been around long enough that people’s viewpoints have been solidified.

Artists feel that they’ve been made obsolete. Art has never been valued, and having to compete with a machine that makes stuff faster, “cheaper”, and without any opposition makes them valued even less by society. People can use as much supportive language to claim it can be worked with, but at the end of the day an artist has been replaced by a machine that does not create from human experience. It might seem like it does, but it doesn’t.

This is the sort of stuff the uncanny valley is supposed to protect us from.

1 Like
77

So, gunna throw my 2 pecos in again. I think it’s important to remember that generative AI is intended to be a tool of convenience. A lot of the usage you see with AI is, indeed, usually just blatant laziness. Not only that, but there are a lot of people that tend to be disrespectful and unethical about their usage of it as well. Not only do the individuals in who foolishly brandish the title of ‘ai artist’ tend to be out right thieves (I do believe that the usage of an artist’s works, without their permission, is objective, irrefutably theft. This is a twofold theft, as they both steal the art for reusage commercially, but they also use the IP of the artist as well without permission, which is also a form of straight up theft.), but they’re also usually hypocrites of the highest order. They will train their AI on artists they like, but then turn around and bash and rally against the artists who they plaiarise from out of angst, jealousy, and the truthful claims of stealing their art. I agree that unethical and commecial usage of generative AI should be treated for what it is.

With all that being said, I do think there is a world where, when artists’ wishes are respected and the use is ethical, generative AI is not a bad thing. Getting permission for usage of the art you use to train a generative ai should become the standard. For those who say no, their art should not be utilized to train a generative model. There could/should also be a possibility of a ‘trainers fee’. Any images generated from an AI, trained on a specific artist’s works, are products of the artists work, even if it wasn’t the artist themself that physically produced the image(s). What happens if an artist wont accept a ‘trainers fee’? Why, that simply becomes the artist not wanting their art used for gen ai, which means we respect the artist’s wishes and don’t utilize their works.

I think the reality of rhe situation is that, even when the Ai speculative bubble bursts, ai is definitely still going to be around and a tool people are going to utilize. If you feel like you don’t want to use it, you are perfectly fine not utilizing it. That being said, you can’t really stop others from using it entirely, so rather than fight that impossible battle, fight the admittedly hard but worthy fight of banning UNETHICAL usage, and promote the ethical and respectful usage instead.

2 Likes
78

I am fully aware .-.

I said it was a dumb strawman in the same comment

79

Another post about AI and yet another pointless discussion. One side telling that AI is bad, another side telling that AI is good. Maybe lets just think from WeightGaming forum perspective? What will happen if neural network generated pictures going to be banned on this forum? Those developers who use this tool will move to another site and post their games there. People who interested in such games definitely will find them anyway. Will this somehow help to stop spreading games which made with neural network tools? No. So, what is the reason to ban AI games on forum? Because some people hate AI? Or maybe some people think this ban somehow help humanity to fight against AI? Why not just make a separate thread for AI games on forum like it is for “Table top games”? People who hate AI won’t use that thread and that is all.

8 Likes
80

You pasted this output straight from an AI chatbot.

2 Likes
81

Re-framing this back to the community is important. I have a genuine question for you. You are a 3d artist and modder which takes a lot technical and artistic skill. After DeviantArt opened the AI art floodgates has it impacted how you use the site and what you get out of it? Do you feel that if you were starting as a beginner you would still bother with DeviantArt?

82

Could you post proof? Are you are accusing it of being ai just because it uses more academic language? That’s not the same. You may think its a pretentious writing style but accusing it as AI written is a serious accusation.

83

The point is mostly that witchhunting ai in a small space like this isn’t really healthy for the community.

There was one dev here who was also an author, if you looked into him further he used realistic~ ai gen’d covers for his stories.
One user then took apart the game and decided the code was too good for it to not be ai, despite the dev saying it wasn’t. Whole thread devolved and the dev quit the site. This kind of behavior just shouldn’t be acceptable.
He is also still developing the game, just not interacting with this community.
There are other cases too.

I get disliking ai.
I get it if you’re an artist and feel you’re losing commissions or having your IP stolen/replicated.
I get it if you feel you put a lot of time into this and feel obsolete.
I get it if you’re concerned less people will learn to draw/create in the future.
I might think it’s a little melodramatic to say the human spirit is under siege.
I might think the environmental complaint about ai specifically is kind of silly when you’re already on the internet downloading/streaming things.

But I get it.
All that’s great.

How does harassing John S. olo’Developer on a niche game dev forum for an even nicher topic help rectify any of that?

My point’s always been that no matter your opinion on it, this space isn’t really the arena for it.

4 Likes
84

I was going to be more diplomatic and reserved on this, but the further I looked the more obvious it became that this is AI.

You only need to look at Jack’s post history to discover that this extensive academic language is entirely out of character for them. By their own words, they’re a native Spanish speaker that finds it overwhelming to read everything in English. Their writing was perfectly understandable, but still “off” enough to know that they weren’t a native speaker. This was a little over 2 months ago: A game...how?

To go from that, to this level of sophisticated academic language in just 2 months? I’m not buying that. Plus the fact it’s complete with the usual AI tells of excessive em dashes, meandering sentences and irrelevant talking points. Then you just look at some of Jack’s shorter comments in this thread and see that they read completely differently.

Then there’s all of the references. Haphazardly thrown in wherever it looks like they’re making a point to make their arguments seem more serious. Except the vast majority of them aren’t referenced properly (after forcing my eyes to not glaze over, this sort of writing would never be accepted in any form of serious academical study), and the papers they reference are completely irrelevant to the topic at hand. I’m not talking about the fact we’re on a fetish forum, these papers have no bearing on AI whatsoever.

Let’s take the first one I saw, a 2014 paper by “Bookchin, M.”. This man is actually Murray Bookchin, who died in 2006, and therefore obviously could not have written that paper. So where did this reference come from? In the top post Jack was kind enough to include a reference list, where we see:

Searching this we find this paper, a social ecology piece written in 1965, uploaded to that website in 2014. Clearly, the AI has simply scanned through a massive database of academic papers, seen that this had the word “technology” in the title, and hallucinated quotes out of it to write Jack’s drivel.

Searching other references brings similar results, with “Bakunin (2013)” likely being Mikhail Bakunin, who died in 1876, and who had a book added to Amazon in 2013. Ironic, given he was a revolutionary anarchist.

I didn’t expect to be spending the early hours of the morning debunking the validity of a fetish forum post, but such are the wonders of the internet. Ironically, Jack seemed to hold AI in high regard in that same post just 2 months ago, claiming game engines other than RPGMaker were too complex for even ChatGPT to handle. It does make me wonder why they decided to go on this anti-AI crusade in the first place, and especially while using AI to do so.

Oh right, of course, this is the internet. It’s ragebait. And we’ve all lost.

9 Likes
85

Oh right, of course, this is the internet. It’s ragebait. And we’ve all lost.

At least this was a sensible chuckle

1 Like
86

It didn’t impact me personally because of how I am using deviantart: I subscribed to a lot of artists and a few groups there and so by using “deviants you watching” tab I am getting only posts from those artists and groups. Yes, a lot of people who generate AI started to post in groups which I am watching too, but since all content from one artist gathering in one “window” on deviantart if I see AI art I can just remove whole window to not bother with checking every post made by this person. I subscribed only to two persons who post only AI pictures, because I think they generate enough quality content which I can use later as example for models sculpting.
Since I am 3d modeller who mostly known as creator of modifications for video games currently AI can’t provide anything which can replace my skills, but I personally would be very glad if AI could generate good 3d models, because I create a video game too and wanted to use it as tool to speed up development. Currently AI can generate pretty good 3d props and even character models, but they are useless for video games, because hard retopology needed (polygons count, uv map and topology is shit). Also, AI models published on other sites since no one going to look for them on DeviantArt.
It is hard for me to answer on your question about starting on DeviantArt as beginner, because my content is very different from 2d artists. I think when you start post your content it is better to post it everywhere possible to gather more subscribers and check statistic by yourself. So, if it was me I would post stuff on DeviantArt, X (twitter), BlueSky, Aryion (if it is vore related content) and Furaffinity (if it is furry related content) at the same time. Of course in my profile on all this sites I would add links to all sites where I post content to let people know that they can watch it wherever they want.

87

Thanks for substantiating the claim.

Regarding the specific Bookchin paper which I spent some time reading the best faith interpretation for me would be that this got conflated with the wrong source date of when it was posted to the website, but the mistake adds credence to your claim.

Another tell is that I can’t find an argument quite like what Jack made in the paper. Bookchin argues for the automation and elimination of certain labor like mining and suggests a technological shift adapted to social life revolving around a more communal and artisanal form rather than the way we currently structure our lives to meet the technology designed to meet the needs of global mass production. Bookchin argues to use technology for automating toil and enhancing the spiritual and artistic elements of work. Not a man being an appendage of machinery creating standardized products. I do think the argument is applicable to AI but Jack doesn’t convey it that well, and given the lack of solid quotes and the mediocre parsing lends more credence to your argument.