And the games are getting pretty good.
One of the most exciting communities in game dev right now is one that most people pretend doesn't exist. Nobody's sharing their work, because they used AI. In an industry with real concerns about job displacement and creative integrity, that's enough to get you written off.
I joined the AI Game Dev Org Discord server after using Claude Code to develop my own game. I expected to find a mountain of slop. Instead, I found a quiet community of developers sharing their work, playtesting each other's games, and giving each other honest feedback.
There's something freeing about being on the outside. These devs don't need to fit into anyone's box. They're not trying to prove AI is the future or win an argument online. They just make games for the sheer pleasure of making games.
This community reminds me of what we tried to build at Rec Room, where I spent six years as a software engineer: a place where people just make things for the fun and love of it. Last week I played through a bunch of games from the server. None of them are ready yet. But a few had real charm.
The Games
Agent Arena
Shmup Golf is a small side-scrolling shooter written in 10 lines of code. You pilot a ship, weave through enemy bullets, and shoot back. The developer's goal is minimizing program size for the fun of it, like code golf meets game dev. The console aesthetic is cool and it feels like a legit little game. With some difficulty scaling it could really click.
I recommended they get the game running in the browser and they were able to do so successfully.
Closing
AI hasn't taken over game dev. These games make that pretty clear. But a human with taste plus AI can get all the pieces in place fast. What's missing right now is scale and the design sense to make everything fit together. That gap is closing. I think someone in a community like this is going to make something really special.
Read the original article
Comments
I am excited about game dev with AI, but the games you posted are kind of a joke.
My kids made similar games with Claude code in js.
Was hoping to see some serious indie games, but these looked pretty terri-bad.
Is anyone building the next SimCity, Civilization, etc.?
By tyleo 2026-03-0313:18 I largely agree with you, but that's honestly the part I like. Just like your kids, there are folks here getting value out of something that wasn't accessible to them before. My most cherished artwork isn't the tasteful stuff, it's the crayon drawing my niece made for me on my wedding day. The low-stakes nature of this AI content feels similar; people are doing it for the sheer pleasure and aren't afraid of meeting anyone's bar. A lot of it is noise right now, but I suspect we'll see it develop into something really interesting if the pattern continues.
By vunderba 2026-03-036:49 Shameless plug - and nothing so grandiose as SimCity but I built a pretty substantial 2D/3D blindfold trainer chess game. It's by no means "vibe coded" though, and there's a fair bit of manual work around the 3d modeling that I had to roll myself.
Even with that I'd still say 70% of the code was written using LLMs with the opencode agent.
By RobertLong 2026-03-037:44 I spent quite a bit of time making games/social WebXR stuff in Three.js. There is unfortunately a ceiling to what commercial success you can have there. Sure there's all the .io games which see a bunch of kids playing games from their school computers, but as much as I'd like to see it, web games aren't going to be as impressive as native PC/Console games. It's not really technology, mostly just the market isn't there.
The models are pretty decent at building simple Three.js/Phaser games, but if you want to work with Unity/Unreal/Godot you're going to need a MCP or other tool to get them to work with the engine's tooling/context. I just so happen to work on one for Unity https://bezi.com
I will say, while I think the current models are very impressive with generating code for most game mechanics. They are still terrible at spatial awareness. Gemini Pro 3.1 is showing some promise here, the latest Opus/Sonnet models are...ok. But there's still a lot left to be desired. You also still really need to know how to make games both creatively and technically to pull off prompting a game into existence.
So are you going to vibecode your way to the next SimCity / Civ without knowing some game dev? Probably not right now and I think that's for the best. People want games that are creative and unique. But a passionate hobbyist who has never made a game, knows some programming, and has a vision for a great game now has an amazing tool set to build their dream game and that's pretty cool!
By whattheheckheck 2026-03-041:20 There is a civilization ai game funded by ycombinator
My latest game BossBattle[1] (html5, have a play) uses AI for graphics, some strobe effects, a C64 loading screen shader.
I have decided to lean in to it and I will document all the places I use AI in the game on my blog[2]. Not everything works, notably 3D assets[3] and sound effects.
There is a lot of human content… i paid for a lot out of my own pocket and have limited budget. It started in 2021 before chatgpt. LLMs cannot do everything and that’s not the purpose.
Generative AI makes me as an solo indie dev able to make the game. Without the AI the game wouldn’t exist
[1] http://epicwin.team/play/solo/BossBattle/ - (public beta) .
[2] https://generative-ai.review .
[3] https://generative-ai.review/2025/08/3d-assets-made-by-genai...
I commend you for being willing to share.
The "able to make the X" thing is something haters love to ignore.
Best of luck with your beta!
By bigyabai 2026-03-038:53 I don't think it's hate; the games market was oversaturated before AI, and it's even more saturated with it. Players have limited time and patience that usually doesn't work out in-favor of indies.
By krapp 2026-03-0311:00 Making the game that OP made, a basic Simon-type color matching game, isn't that difficult. None of the vibe-coded games I've seen posted to HN would be difficult to make in any engine. If that or any other such game had been written by a human being and posted here it would be getting a similar reaction because none of them are any good on their own merits.
But because they're AI, and only because they're AI, people gas these things up regardless of quality, almost in spite of it. That's why people are mad. Not because they're "haters" but because people are flooding HN and the rest of the web with slop and insisting on being treated like masters of their craft.
And again, if a human had written this game in its entirety and posted it here, it would get flagged as spam, because it is first and foremost a bad game. Even as a "beta." And OP can't improve on it much because they're limited by whatever the AI happens to generate. Even just giving them encouragement is pointless.
"Games are impossible without AI" is a paltry excuse. My nephew, who isn't even in high school, is learning coding in school and is making games in Roblox. It's never been easier. If it's true that making a game in AI takes actual skill and work, as people claim, then making it in an engine is not at all beyond anyone's grasp.
If y'all want to shut up the "haters," make something with AI that's actually good, a game with a coherent visual look, well designed gameplay, an interesting concept, that's at least as good as a tutorial project for any existing game engine.
By tyleo 2026-03-0312:29 This is the exact sort of thing being unlocked that inspired me to write the article.
I’m so happy folks are finding fun and creativity in this space.
By fatata123 2026-03-0313:54 [dead]
By yuppiepuppie 2026-03-037:032 reply If you want to see what the HN game dev community is building, I’ve been curating the list here (https://hnarcade.com) for the past months.
The output is quite impressive. And having spoken to a number of the developers, it does seem like AI has had a massive impact on delivering their ideas.
By mcteamster 2026-03-0311:29 Awesome, I've submitted a game I made last year (https://blankwhite.cards) which was itself a remake of a game I made in 2019.
In my experience it took 5 years to find the time and motivation to build a v2.0 by hand. AI has since accelerated the production process to help me ship features that would otherwise have taken me another few years to even consider doing.
By spppedury 2026-03-037:25 [dead]
