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

[Jack725]

Hi,

I’ve worked in this medium long enough to see how the passion for creating interactive worlds has transformed into a competition to produce faster and cheaper. I should clarify that I’ve never made a video game; I’ve only created stories, characters, and dubbed into Latin American Spanish (my language; if this is poorly translated, it’s Google Translate’s fault). Today, faced with the expansion of images generated by artificial intelligence (AI), the video game industry faces a decisive ethical and aesthetic dilemma: will we continue to allow machines to replace human art, or will we defend the creative soul that gave rise to this medium? My position is clear: we must definitively banish the use of AI-generated images in video games. This is not a fear of progress, but an act of cultural defense against a technology that threatens to empty our art of meaning. We are not EA, we are literally the proletariat and if as a community (however small) we do not start to put limits on garbage content, the desire for meaningless production, purpose and in general, those few feederism games that have soul, personality and depth… will get lost in the shit creating a market of overproduction where being a good creator will not be profitable or appealing.

Images produced by AI are not art, merely a mimicry of it. The machine does not imagine, remember, suffer, or contemplate; it only averages. Every pixel it generates is based on millions of past human examples, but devoid of intention. (Anyone who knows about art will know why the fact that art lacks intention already invalidates it as art.) Recent studies have shown that humans distinguish and value works they believe were created by people more than those made by AI, even when they cannot visually differentiate them. In the study by Bellaiche et al. (2023), participants rated human works as more beautiful, profound, and with greater creative merit. This demonstrates something essential: artistic value lies not in the aesthetic result, but in the consciousness behind the act of creation. A video game, as a complete work, cannot be sustained by a mimicry of humanity. Furthermore, using AI in visual production leads to an alarming homogenization (there are no products that look new, there is no innovation, they are all the same). Generative models are trained on millions of pre-existing images, and therefore tend to reproduce the same visual patterns, the same palettes, the same gestures. What appears to be diversity is actually statistical repetition. The result is an ocean of visually indistinct titles, devoid of identity or risk. According to Koivisto and Grassini (2023), humans still outperform AI in divergent creativity tasks, demonstrating a much greater capacity to generate original ideas. Relying on AI to create the worlds we imagine means giving up the very possibility of aesthetic innovation.

Another aspect that should concern us is the precarization of creative work. Every time a company decides to replace an illustrator or designer with an image generator, it’s not just saving costs: it’s contributing to the erosion of the artistic profession. And you, as the offspring, are you going to be part of that movement? AI is trained on millions of human works taken without permission or compensation, expropriating the efforts of artists around the world to produce a diluted copy of their style. This isn’t progress; it’s digital extractivism. As the Krisis Group (2002) warns, contemporary capitalism seeks to eliminate human labor while maintaining the ideology of performance. AI embodies that contradiction: it promises efficiency at the cost of erasing authorship, ethics, and the dignity of the creator. Those of us who defend creativity as an act of freedom cannot accept this substitution. AI doesn’t liberate the artist; it replaces them. Murray Bookchin (2014) warned that a technology is only liberating if it expands human capabilities, not if it replaces them. In this case, the developer ceases to be a creator and becomes a prompt operator, a recombination technician. In cultural terms, this amounts to a regression: from the work to the formula, from gesture to instruction.

The consequences are also felt in the game design itself. AI-generated titles tend to focus on superficial visual stimuli—changing models, characters gaining weight without an understood anatomy, multiplying effects—but with empty mechanics. This type of product confuses aesthetics with experience and transforms the player into a passive consumer. There is no longer challenge, exploration, or discovery (in terms of experience creation), only repetition. Instead of being interactive art, they become dopaminergic displays, designed to maintain attention without offering any meaning. As Berardi (2023) points out, we live in an era in which symbolic labor has been absorbed by automation, producing an excess of signs without soul or context. AI-created video games are the clearest manifestation of this saturation.

Even more serious, AI-generated images tend to be culturally biased and, depending on the model, have political objectives that censor their true capabilities. The datasets used to train them are dominated by Western, commercial, and masculine aesthetics, resulting in uniform representations devoid of real diversity. Thus, AI not only replaces the artist but also imposes a global visual canon that erases identities (anyone who understands sociology and cultural intersectionality should be shitting their pants right now). Allowing its expansion into video game development would be tantamount to accepting a form of colonization. We can no longer normalize the cannibalization of human works to produce soulless content. We must legislate, agree, and, above all, commit as a community of developers to ban the use of AI-generated images in video games. Art doesn’t need to be faster, but more human; it doesn’t need to be more perfect, but more truthful. If we accept this substitution, we will lose not only jobs, but also the very meaning of creating. And when that happens, there will be no more video games, only playable code produced by algorithms that have never known what it feels like.

Bibliography (yes, i am a nerd):

Bellaiche, L., Shahi, R., Turpin, M. H., & et al. (2023). Humans versus AI: whether and why we prefer human-created compared to AI-created artwork. Cognitive Research: Principles and Implications, 8(42). Humans versus AI: whether and why we prefer human-created compared to AI-created artwork | Cognitive Research: Principles and Implications | Full Text

Berardi, F. (2023). Half a Century Against Work. Bifid Canon. Autonomous City of Buenos Aires: Tinta Limón.

Bookchin, M. (2014). Towards a Liberating Technology. The Anarchist Library.

Grupo Krisis. (2002). Manifesto Against Work. Barcelona: Virus Editorial.

Koivisto, M., & Grassini, S. (2023). Best humans still outperform artificial intelligence in a creative divergent thinking task. Scientific Reports, 13(14501). Best humans still outperform artificial intelligence in a creative divergent thinking task - PMC

[BadIdea4]

[jeffskeep]

I generally agree, mostly because ai generated content just isn’t interesting.

But I’m gonna hit you with a dumb strawman anyway just because:

Did you know google translate is AI? It’s also a transformer model using a neural network.

Why didn’t you pay a human interpreter to translate this for you?

[Jack725]

Hola,

Llevo trabajando en este medio el tiempo suficiente como para ver cómo la pasión por crear mundos interactivos se ha transformado en una competencia por producir más rápido y a menor precio. Debo aclarar que nunca he creado un videojuego; solo he creado historias, personajes y doblado al español latinoamericano (mi idioma; si está mal traducido, es culpa del Traductor de Google). Hoy, ante la expansión de las imágenes generadas por inteligencia artificial (IA), la industria de los videojuegos se enfrenta a un dilema ético y estético decisivo: ¿seguiremos permitiendo que las máquinas reemplacen el arte humano o defenderemos el alma creativa que dio origen a este medio? Mi postura es clara: debemos desterrar definitivamente el uso de imágenes generadas por IA en los videojuegos. No se trata de miedo al progreso, sino de un acto de defensa cultural contra una tecnología que amenaza con vaciar de significado nuestro arte. No somos IA, somos literalmente el proletariado, y si como comunidad (por pequeña que sea) no empezamos a poner límites al contenido basura, el deseo de producción sin sentido, de propósito y, en general, de esos pocos juegos de feederismo con alma, personalidad y profundidad… se perderán en la mierda, creando un mercado de sobreproducción donde ser un buen creador no será rentable ni atractivo.

Las imágenes producidas por IA no son arte, simplemente una imitación de él. La máquina no imagina, recuerda, sufre ni contempla; solo promedia. Cada píxel que genera se basa en millones de ejemplos humanos pasados, pero carece de intención. (Cualquiera que sepa de arte sabrá por qué la falta de intención ya lo invalida como arte). Estudios recientes han demostrado que los humanos distinguen y valoran más las obras que creen creadas por personas que las creadas por IA, incluso cuando no pueden diferenciarlas visualmente. En el estudio de Bellaiche et al. (2023), los participantes calificaron las obras humanas como más bellas, profundas y con mayor mérito creativo. Esto demuestra algo esencial: el valor artístico no reside en el resultado estético, sino en la consciencia que subyace al acto creativo. Un videojuego, como obra completa, no puede sostenerse imitando a la humanidad. Además, el uso de IA en la producción visual conduce a una homogeneización alarmante (no hay productos que parezcan nuevos, no hay innovación, todos son iguales). Los modelos generativos se entrenan con millones de imágenes preexistentes y, por lo tanto, tienden a reproducir los mismos patrones visuales, las mismas paletas, los mismos gestos. Lo que parece diversidad es, en realidad, repetición estadística. El resultado es un océano de títulos visualmente indistintos, carentes de identidad y riesgo. Según Koivisto y Grassini (2023), los humanos aún superan a la IA en tareas de creatividad divergente, demostrando una capacidad mucho mayor para generar ideas originales. Confiar en la IA para crear los mundos que imaginamos significa renunciar a la posibilidad misma de innovación estética.

Otro aspecto que debería preocuparnos es la precarización del trabajo creativo. Cada vez que una empresa decide sustituir a un ilustrador o diseñador por un generador de imágenes, no solo ahorra costes, sino que contribuye a la erosión de la profesión artística. Y tú, como descendiente, ¿vas a formar parte de ese movimiento? La IA se entrena con millones de obras humanas tomadas sin permiso ni compensación, expropiando los esfuerzos de artistas de todo el mundo para producir una copia diluida de su estilo. Esto no es progreso; es extractivismo digital. Como advierte el Grupo Krisis (2002), el capitalismo contemporáneo busca eliminar el trabajo humano manteniendo la ideología del rendimiento. La IA encarna esa contradicción: promete eficiencia a costa de borrar la autoría, la ética y la dignidad del creador. Quienes defendemos la creatividad como un acto de libertad no podemos aceptar esta sustitución. La IA no libera al artista; lo reemplaza. Murray Bookchin (2014) advirtió que una tecnología solo es liberadora si amplía las capacidades humanas, no si las reemplaza. En este caso, el desarrollador deja de ser un creador para convertirse en un operador puntual, un técnico de recombinación. En términos culturales, esto equivale a una regresión: de la obra a la fórmula, del gesto a la instrucción.

Las consecuencias también se dejan sentir en el propio diseño del juego. Los títulos generados por IA tienden a centrarse en estímulos visuales superficiales —modelos cambiantes, personajes que ganan peso sin comprender su anatomía, efectos multiplicadores— pero con mecánicas vacías. Este tipo de producto confunde la estética con la experiencia y transforma al jugador en un consumidor pasivo. Ya no hay desafío, exploración ni descubrimiento (en términos de creación de experiencias), solo repetición. En lugar de ser arte interactivo, se convierten en exhibiciones dopaminérgicas, diseñadas para mantener la atención sin ofrecer ningún significado. Como señala Berardi (2023), vivimos en una era en la que el trabajo simbólico ha sido absorbido por la automatización, produciendo un exceso de signos sin alma ni contexto. Los videojuegos creados con IA son la manifestación más clara de esta saturación.

Más grave aún, las imágenes generadas por IA tienden a presentar sesgos culturales y, según el modelo, tienen objetivos políticos que censuran sus verdaderas capacidades. Los conjuntos de datos utilizados para entrenarlas están dominados por la estética occidental, comercial y masculina, lo que resulta en representaciones uniformes carentes de diversidad real. Por lo tanto, la IA no solo reemplaza al artista, sino que también impone un canon visual global que borra identidades (cualquiera que entienda de sociología e interseccionalidad cultural debería estar cagándose en los pantalones ahora mismo). Permitir su expansión al desarrollo de videojuegos equivaldría a aceptar una forma de colonización. Ya no podemos normalizar la canibalización de obras humanas para producir contenido sin alma. Debemos legislar, acordar y, sobre todo, comprometernos como comunidad de desarrolladores para prohibir el uso de imágenes generadas por IA en videojuegos. El arte no necesita ser más rápido, sino más humano; no necesita ser más perfecto, sino más veraz. Si aceptamos esta sustitución, no solo perderemos empleos, sino también el sentido mismo de crear. Y cuando eso suceda, no habrá más videojuegos, solo código jugable producido por algoritmos que nunca han conocido la sensación.

Bibliografía (sí, soy un nerd):

Bellaiche, L., Shahi, R., Turpin, M. H., & et al. (2023). Humans versus AI: whether and why we prefer human-created compared to AI-created artwork. Cognitive Research: Principles and Implications, 8(42). Humans versus AI: whether and why we prefer human-created compared to AI-created artwork | Cognitive Research: Principles and Implications | Full Text

Berardi, F. (2023). Half a Century Against Work. Bifid Canon. Autonomous City of Buenos Aires: Tinta Limón.

Bookchin, M. (2014). Towards a Liberating Technology. The Anarchist Library.

Grupo Krisis. (2002). Manifesto Against Work. Barcelona: Virus Editorial.

Koivisto, M., & Grassini, S. (2023). Best humans still outperform artificial intelligence in a creative divergent thinking task. Scientific Reports, 13(14501). Best humans still outperform artificial intelligence in a creative divergent thinking task - PMC

En español por gracioso

[Jack725]

I understand what you’re trying to say and;

1. It wasn’t AI before
2. False equivalence fallacy with a bit of the scarecrow fallacy

[jeffskeep]

Yeh, that’s why I said it was a dumb strawman (scarecrow)

[Jack725]

Sorry im autistic. I did not undertand your joke (and yes, i speack english but is more easier with a Google translate help).

[drxprime]

Google Translate has always been a form of AI… AI isn’t suddenly new as of 3 years ago.

For commercial mass-media projects, I agree with you. At least for things solely generated with AI, even for social posts and such (those are the worst). It is hard to remove it completely from the tool-chain in every facet of production, is Grammerly running on the game dialogue an issue? How about an artist which content-fills a texture to manipulate it? Like there’s going to be some level of tooling which uses some form of AI underneath (LLM or not) which is going to get used somewhere inevitably, whether everyone on the team knows it or not.

For the random fetish game on the internet, meh. At least for images, I’ve seen it inspire more people to write and create than not. Yes, it’s not as high-quality, and it should be disclosed as such. AI bots/writing is generally a lot lower quality of an experience I’ve found so far, not worth the time to play. That’s part of just having a market and reviews is about, people can choose what to consume. They’re free to do so. Just like others don’t have to listen to your rant and move on too.

It’s hard to impose limiting a freedom, especially something like this which can be done locally by anyone. The cat’s out of the bag. As much as I may hate aspects of it, we kind of have to figure out how to move forward. But it should at least be transparent, and I hope once the bubble bursts (since really no one actually makes money with this junk yet and no sane people really want it) that the demand for human content comes back and hopefully we’ve sorted out how to restore some balance to the internet… there’s a lot bigger problems in the world right now than the prevalence of AI art for porn.

[Ares42]

While a lot of your points are entirely valid I feel they are somewhat misplaced. Absolutely large studios/professionals/ those that have the resources should absolutely invest in and promote originality and human creativity. I’m all for that. That being said I feel that the use of AI has it’s place for those of us that are not professionals. The hobbyists that have a creative vision but lack the resources to hire artists and what not so they can make something. I’ve wanted to make games since I was a teenager but practical considerations of getting a job that paid the bills took precendence and life did that thing it does and getting into that career path somewhat passed me by.

Then AI came along and suddenly there’s an opportunity for people like me that have lives and family commitments and careers to make games as a hobby and see it taking shape. There’s still a learning curve and scope to improve but we don’t have to get frustrated and give up when trying to write lines of code as non programmers and feeling like we’re getting nowhere.

And for me at least a LOT of work goes into what I do with AI. I don’t just let it spit out any old crap and go “that will do”. I often do post processing work on it in art software too. I’ve seen a lot of games that use AI art and I do agree it’s like they all use the exact same model and art style. Speaking for myself, I try to do it differently.

Would it be better if I could make artwork for my game without AI? Absolutely. But I’m a pragmatist. I can only devote so much time to anything. And at the end of the day it’s a tool like anything else. I know I’m not going to sway the die hards that just think AI is pure evil (insert joke about terminator/the matrix here if you like) but to anyone on the fence or no strong opinions I would say this: Most of us here don’t make games professionally or ask for money upfront to play these games. Yeah, some of us have patreons and other donation systems for people that enjoy our work to contribute in what I see as a show of appreciation - but generally speaking our content is free and created just for the fun of making something. So when it’s all in fun, and there is no genuine harm done…just enjoy yourself. Whether it’s AI or not - at this level it doesn’t really matter in the grand scheme of things.

Like what you like, enjoy what you enjoy. And don’t dismiss something out of hand just because it uses AI.

[Axxis]

As soon as I see AI in a project I immediately tune out as there is clearly no soul in the project.

[coldsteelj]

I fully accept neural network use in projects, but I personally don’t play in such games. The only thing which I want to see on this forum is a complete separate section for games where main graphic was generated with neural network. Actually, I think would be better to have separate thread for AI games and separate thread for Visual Novel plus Text Based games. I know that we can navigate by using TAGs, but separate threads sounds more useful for separating normal games from something story related without actual gameplay. I would be glad to someday open only one thread and check all projects there which I make sure have some gameplay instead of reading tons of posts with Visual Novels and text games to try not to miss out on any interesting projects that are buried among all this.

[Dapk_Macmep]

I generally agree, I prefer it to help, and not deprive my life of meaning (creativity, for me, is what makes you go to work/study, and then come back to do what you like, and not just lie in bed asking the question “why should I go anywhere? why do I need money if I don’t really live, like a machine, I just perform my function.”), I heard about an AI that can create games (I haven’t tried it, and screw it) and it’s sad, I gave up seriously using AI (creating images, music and code, to say that I am a neuro artist / musician / programmer, although in fact, such people are not capable of anything without their boxes), but if I can barely do images and music, but I can do something, but with programming I will have to turn to both AI and people who at least a little understand how to write code (a code-free option for creating games is good, of course, but I can’t limit myself like this forever.)

[Ares42]

That’s a gross generalisation plain and simple.

[madone]

Totally disagree with your assessment. It will speed up the project development process while keeping costs down. It’s just another advancement of technology to be used in things we haven’t yet seen how far it can go. As more and more people have access to this technology it will promote more creativity in more people. This will become more apparent especially in video media, which already has seen a growth in independent movie producers on a shoestring budget.

[Wulf_Diomhaireachd]

You know you can add certain tags to be ignored through your account settings? Outside of the odd post that forgets to add the right tags (in which case moderators or regulars like myself will add them soon after), you can block all of those posts tagged with ai or visual-novel or whatever from appearing.

There is a way to link tags in messages, I just can’t remember how.

[coldsteelj]

If I wrote that I don’t play in such games that doesn’t mean I don’t read those posts. There is always a chanse that someone added text-adventure tag, but there is some combat minigames involved and even characters visuals. I personally would be insterested in such project and so I check every post to see if there is something unique which I may like. Also, I guess for new users could be troublesome to find that they can go to profile settings and in “tracking” block specific TAGs. Some users may create a post without adding needed tags. I asked users to add specific tags to their games multiple times. So, since we already have a section specifically for “Table Top Games” I think could be nice to have another one for “Visual Novels” too or “AI” involved stuff (since people start arguing about AI again and again). But of course, I don’t insist and just decided to express my opinion
P.S. If you want to see only posts without specific tags you need to go to profile settings and add those tags there, but if you want to check posts with this tags again you need to go back to setting and remove tags. Well, pretty time consuming functionality. When I use search I can filter posts by tags, but I can’t unblock my blocked tags in the process without going back in profile settings. Or maybe I just can’t find such option when I use “search” on forum? That one which let me not only see posts with specific tags, but to not see posts with specific tags too.

[hanzsolo1]

Not everyone has the time, money or capability to make actual art the way i see it if you don’t like it don’t bother playing it. Calling to ban AI art out right severely limits the capability of the community to make anything beyond text based games and the way i see it that just either A kills the community cause theirs hardly any content in this fetish game site B forces people to have to pay someone from the community to work on their stuff which can lead to many issues that i wont even bother to list honestly the whole ban AI thing is dumb why because some art snobs don’t like it? now I’m not saying its acceptable for some huge studio with resources to outsource to AI no that’s not right BUT the content here isn’t intended to go up in some gallery like you said we aren’t EA the majority of people here on this site don’t have the capability to code or let alone do art honestly this whole debate is stupid if you don’t like it don’t use it but don’t try to limit the community as whole because of your preferences for content i have my preferences and things id like to not see at all regarding some of the content here and I’m not calling to ban it. also i saw cold steel mention something about separate threads of sorts for AI stuff I’m personally indifferent to that however if that’s added to separate Conventional work from AI when tags do a much simpler version of that (which in some cases tags aren’t added and either way how would X post get to the right thread in the first place) Then the admin team will have to further extend that resource to other things like furry, vore or giantess or this game isn’t weight gain its only breast expansion or this and that BUT I’m getting off topic if you see its done with AI just ignore the best thing i can say to do is better tag enforcement and as another user stated like 3 times you can disable what tags you want to see and what not

[FA_Gamer]

I think my major disagreement with OP is banning AI in videogame development is gatekeeping videogame development for people with no art skills or money to hire an artist.

I do agree however that projects with capitol should pay artists instead of using an AI image generator. (I think the large majority of projects on this site don’t have development money)

That being said AI is just a tool and can be used to make art. If your not familiar with Anime Rock Paper Scissors you should look it up. Corridor Digital trained their an AI on open source anime to get a style then fed the AI video they shot of themselves to generate the video in that style. They used AI as a tool to create a unique piece of Art. Yes they did a lot of other work without AI to make it look good but if you blanket ban AI you have to ban projects like this that use it as a tool.

[sypern]

honestly i agree completely, if your excuse is that you “don’t have the money” then either make a text-based game, or learn to draw, talent has nothing to do with it, its a learned skill, and if you aren’t willing to put in the effort, then your “game” isn’t worth my time. but the main issue that people are talking about is GENERATIVE ai, not the type used in google translate, or the ai that is used to make enemies behave the way they do.

[Ares42]

What about people that can’t draw? Not because they don’t have the time to devote to learn the skill but, who, no matter how much time they devoted to it could never draw due to limitations that have nothing to do with skill or time - i.e. physical or neurological conditions? We can’t draw - but we can type or use text to speech. Tell me where the lack of effort is for those cases.