Some people call it bad news for the gamedev ecosystem but I definitely call it good news for game developers. AI will not take your job, it will get you there 10 times faster. I made this game with pixelfork.ai in 2 hours. Images created with Gemini, then transformed into 3D models with Krea AI, then imported into Pixelfork. SFX was created with ElevenLabs. Everything is AI. No manual code, no manual generation. I'm not sayin this is the way developers should publish their game, but this is definitely the way developers should prototype. Have an idea, just say it, it creates. We are closer than ever. Winter is here)
krapp•4w ago
It looks like one of those fake games they advertise in mobile ads. It's simplistic, repetitive and dull. It's buggy (I died at the start more than once, before it even loaded) and some of the levels seem impossible to finish. And why is the "train" clearly a bus?
This isn't even a good example of a prototype because no one would bother prototyping this.
Show me an AI generated game that's actually decent as a game and not as a tech demo for the AI behind it.
I'm not dumping on this just because it's AI. Had this been coded entirely by hand I would have the same criticisms. If AI generated content is going to compete against human created content then it needs to be judged on its own merits, not by how cutting edge the AI used is.
fkhasiyev•4w ago
Did you read the caption clearly? I said I created this for just a prototype. there can be an errors, it's not a game created till the finest details. just a tech demo. if I spend more time on this, I clearly will make it better and better. Making bus not the train, is just a artistic choice. yes, you are right that sometimes it's buggy, but its happening on the first play after loading screen. then its working good.
its just a beginning, through the end of this year, there will be a lot better ai models capable of creating beyond tech demos.
krapp•4w ago
I'm sorry, but I've seen "finished products" from AI game devs that don't look much better.
This is a simple prototype (so simple I doubt anyone would bother) but the more complex it becomes, the less fine control you'll have, the more you'll have to just hope the LLM gives you something approximating what you want.
And then what? Use what's likely to be complex, buggy and inefficient code that you didn't write and don't understand? Build on top of that? Just keep prompting the AI and have it build the final product as well?
I really don't think this is better than existing methods even for prototyping because AI by its very nature just approximates and that takes creative control away from the developer and the artist. I don't think that's worth a fast turnaround.
fkhasiyev•4w ago
People were talking like this about Cursor, Windsurf and any Code agents 2 years ago. It's improving day by day. It's just a sneak peek now, next year this day, there will be 10x better versions. keep an eye on AI Game Dev in 2026.
foxtrottbravo•4w ago
AI created this, and it shows - but yes I agree somewhat for a rough prototype or experiment this might be enough
fkhasiyev•4w ago
yea. games should not be published like this, it should have final human touch after nailing all the game mechanics, and loop. I have friend who spent 15K USD on just creating a boring Rogue like Tetris game. It looked fantastic but boring to play. game failed. prototyping fast is the crucial point.
6510•4w ago
There is a lot of talk about AI data centers energy use but playing this game my laptop fan took things to levels I never hear before. If enough good enough software is <s>written</s> generated the power consumption is going to be amazing.
It also took forever to load but the game was good for the low 2026 standards.
fkhasiyev•4w ago
krapp•4w ago
This isn't even a good example of a prototype because no one would bother prototyping this.
Show me an AI generated game that's actually decent as a game and not as a tech demo for the AI behind it.
I'm not dumping on this just because it's AI. Had this been coded entirely by hand I would have the same criticisms. If AI generated content is going to compete against human created content then it needs to be judged on its own merits, not by how cutting edge the AI used is.
fkhasiyev•4w ago
its just a beginning, through the end of this year, there will be a lot better ai models capable of creating beyond tech demos.
krapp•4w ago
This is a simple prototype (so simple I doubt anyone would bother) but the more complex it becomes, the less fine control you'll have, the more you'll have to just hope the LLM gives you something approximating what you want.
And then what? Use what's likely to be complex, buggy and inefficient code that you didn't write and don't understand? Build on top of that? Just keep prompting the AI and have it build the final product as well?
I really don't think this is better than existing methods even for prototyping because AI by its very nature just approximates and that takes creative control away from the developer and the artist. I don't think that's worth a fast turnaround.
fkhasiyev•4w ago