Indie Game Development with AI | Hacker News Reader

Indie Game Development with AI

sandromaglione.com

Personally, I don’t care how a game was created (vibe coded or coded by hand) as long as the game is fun. I think it’s cool that LLMs let you focus on making the game fun rather than spending your time just getting the game to run. I’m obviously in the minority on Hacker News though.

It's nice to read. I've been developing games with ai but putting a lot of hard work into the design and development. I'm afraid to share them with a wider audience because the anti ai sentiment in the community right now is so strong.

> I'm afraid to share them with a wider audience because the anti ai sentiment in the community right now is so strong

Find your audiences and cater to them, haters will always be around and will always hate as loudly as they can.

Who are they to judge the creation process of another person's desires without being involved in any sort of way other than being another human with an opinion on the process used?

The way we were vs who we are now. Be passionate! Offer fun games, entertain your audiences!

The audiences exist, but there's ten of you for every one of them. There is a market for low effort - always was - but AI has saturated it worse than ever.

Use AI for the things that aren't art, but craft. And accept that no one will pay for the craft, so the more things you resign to be craft-not-art, the less reason there will be for anyone to play, let alone buy, your game.

I've yet to see an LLM made game where the prompter actually spent any real amount of "saved" time on making it fun.

Fun is a dime a dozen. You can have an LLM tell you jokes all day if you want to. Casino games are fun too. You almost certainly want something more than fun.

I think this discussion highlights something interesting:

Our society is moving towards a more one-dimensional conception of enjoyment, or in other words, "base fun". Not to say that simple fun is bad, but I think it would be a shame if life enjoyment were reduced to straightforward fun and nothing else.

Meanwhile, I’ve spent the last five weeks poking at an MNA and fluid dynamics solver for Minecraft, of all things. That’s the kind of fun I like. It’s what you’re looking for, isn’t it?

Heavy use of AI, of course. I’d never have gotten anywhere on my own; I’ve got my hands full just playtesting everything.

But you're

* not selling it

* not expecting to get the attention for it that someone would get from e.g. porting Sonic to the Amiga.

That makes all the difference in the world.

It's a game. Fun is the literal definition of what I'm looking for

Yes, but tastes differ as to what fun is, and endless banal dialogue in Claudish would, for most people, wear thin as quickly as smalltalk with chatbots.

Same goes for AI-generated environments that are no more compelling than assets produced with traditional procedural generation techniques.

With that said, in games as elsewhere, there's a huge difference between quickly thrown together AI slop and development processes where AI is judiciously used as a labor-saving device.

AI driven development doesn't allow you fine enough control to really implement anything specific, you're stuck with whatever generalities the model derives from the prompt on top of whatever random changes and bugs it introduces between iterations. One could take care of those things outside of an LLM, but let's be realistic - for the vast majority of people "making the game fun" means adding "make the game fun" to the prompt and calling it a day. This is why the technical implementation by LLMs constantly improves but things like affordance, graphic design, level design etc so often fail.

So you would be right if you only get a single prompt of limited length, no assets, no documents, instructions to iterate until complete, and an hour of wall clock time.

So ...ok... there's a few fable 5.2 demos out there that do crazy things with even that.

But yeah... possibly we're more in agreement than not.

If you write hello world in python or slap some paint on a canvas or blow into a concert flute with no practice; it's painfully obvious that you're not done yet. You have a long way ahead of you to make something that looks or sounds like something.

In the case of LLMs and image generators, you often do immediately get something that looks like something. If you're a "majority of people" you'd be forgiven for thinking that you're done, and that that's all it can do.

But that's not true at all. If you're a "some people", you might just realize that You're not done yet . Not by a long shot. That's just a starting point. Bring in more tools. Work in layers. Write 10000 prompts on a single model and throw the results away, just to get a feel for the latent space. Do it again for 20 more models. Have an LLM do research into algorithms on the half that's less important to you. Have one dig up textbooks and papers on the half that's more important to you. Read them. Ask "Has anyone had this idea before?". Look up those artists and writers. Read their works. Be inspired. Now you might finally have a spec for your model that isn't just tropes. Get to work. Write your first 100000 words and put them aside. Now you might have enough skill to design a plot that holds together. Etc. etc.

And a 100 important things more that either I don't know or hasn't even been invented yet.

Work the problem step by step. ;-)

A new thing is: you can apply your artistic vision without needing a gigantic team to help you. That's very neat!

Of course I wonder, what if a large team did get together? How would they work? What could they build?

AI driven development allows as much control as you're willing to put in. You can have it make single digit changes if that's the amount of control you wanted

If you use AI in your game I will not play it. Indie or not. I'm clearly not alone in this thinking, since the CEO of Epic Games is spreading AI-generated propaganda claiming that AI labels "hurt developers" due to significantly reduced sales.

Everyone is using AI in the gaming industry. What you gonna do now?

Not everyone, quit with the doomer propaganda.

There is in the indie scene a large group of people and studios that refuse to adopt it, and proudly advertise this fact. In fact it might turn out to be the very thing that separates ‘indie’ from everything else. And as a player of indie games, I actively steer away from games who have been created through AI assistance, as far as it is possible to tell.

And there is no way to verify it in any way.

It doesn't really matter because in practice, the people who use AI seem to be the type not afraid to admit it, and the people who don't use AI hate it personally so they are usually honest about it. Unlike commercial advertising, there really isn't a great benefit to lying about AI use.

[citation needed]

Common sense. Except maybe most of the graphical artists, but coders, with common engines like UE? Surely.

Stop playing games. Just like I stopped reading AI generated blogs.

Although that won't happen. There will always be some people passionate to create amazing and fun experiences.

Play older games until there is a robust community to identiy human-made art. Enough of those old games to last a lifetime if society doesn't stop eating its own tail before then.

I was talking about coding, not graphical AI assets.

Play more old games. Classic games. I've been playing a lot of old Nintendo games through and emulator and there's a genuine quality to them not in the slop we see today.

We've created so many cool things in the past that there's plenty of things to enjoy now that weren't tainted with AI.

Certainly not Epic, given that Unreal 6 development workflow is strongly based on AI as announced recently on the Tokio Unreal 2026 event, and will even drop C++ for gameplay, it will only be supported on the core engine.

AI is exposing everyones poor workflow management, and in the gaming industry this not only hurts profit but it hurts pride.

If your workflow sucks, your product sucks - and users can sense this from a distance, AI or no-AI.

There will come a point soon where you won't be able to spot the AI in the workflow. This is already happening all over many other industries - some of which, do have proper, robust software development workflows in place as an entrenched standard.

Game guys will catch up.

That's your call. But I see a deeper problem:

"My plan for Spirit Call was a game like Genshin Impact (原神), with a new character every few weeks."

In other words, their plan was to get really rich with as little effort as possible, copying the essence of a highly profitable and exploitative gacha game.

Exploring new games or any media, to me, is all about avoiding the authors who have that as an objective. And right now, AI use for creative media (and even more AI praise in games, like this article) is a stronger warning sign than bargain bin asset packs.

But, I don't hate AI. It'd be hypocritical of me. I am still the person who got fascinated by the first blurry dogslug, and struggled to get Lua Torch running on my already then ancient GPU so I could play with the DCGAN code and make flowers from the Oxford dataset. Generative AI is damn amazing. It's people who would use it to get rich quick who suck.

It's definitely nice that AI can help speed up some chores of the indie development part that use to take so long before. The sad part is that it's used to flood the current indie markets with slop, either with fully automated crap, or with attempts to steal whatever idea becomes viral to make a quick buck destroying a previously good marketplace. It's a shame that some areas of gamedev went from a place of passion, to a place full of grifters.

I am confident that people using AI to come up with unique and creative ideas will raise above all that slop eventually.

It's only a matter for them to embrace AI as well, learn to use it as a tool (and not a replacement), and then indie devs will thrive above the slop.

Get your hands dirty and build something that's what it's all about.

I think games like art should be one of the few areas that aren't developed by AI, because human expression is the point.

I mean code, reports and other stuff, those have utility outside of the creation process itself (although the creation process still has value) but the whole point of media is human sharing.

> I think games like art should be one of the few areas that aren't developed by AI, because human expression is the point.

Being human as a perk: it lets everyone think and do as they please.

People let you think as you please, let them do as they please.

Why are you(or anyone) so gung-ho on restricting someone else's access to creating a game when you are not involved in their personal lives whatsoever?

Create away. But if you try to sell it, I will assume you are a get-rich-quicker, and I'll warn people against your game like I warn people against MLM schemes or 3d printed tat hawked on Etsy.

Like the other poster said. You're free to make your game using AI, and I'm free to not use it, discourage others from using it and boycotting and protesting you for creating it.

I really like the inked town game

Code is creative output, and delegating it to LLMs is at the cost of that creativity.

You won't know what you're missing or what lessons you would be learning. Sure, you're going "fast", and you'll ship something, but it doesn't look like meaningful engagement with the craft or an audience if a developer didn't execute on it themselves.

I also read peoples posts about spending tens of thousands of dollars on local hardware that they could have hired an indie developer to work with, or throwing untold sums at online LLMs that also equate to a year or more of paid development time without LLM use.

On style: the rationalization of LLM use reads like it was written by a LLM, which is kinda offputting but it also smells like highly agreeable output.

> And still, making those games has been hard and intense. Like never before. For weeks.

Try years. Weeks doesn't give you enough time to sit with it. To self-doubt your own arguments for gameplay decisions you made, only to accept them. To run the gamut from loving an asset to hating it to loving it again all while avoiding nuking the project with 1000 tiny self-conscious edits. There is such a thing as moving too fast.

At any rate, I hope the game is fun.

The Inkline UI is so obviously designed by AI - the floating chips with drop shadow stuffed with numbers.

And the author admits they also completely lifted the art style off a tweet.

So I'm not sure where they exercised the 'taste' they mention so many times in the article.

tks for the read, I like the style of your inked game

I honestly think that we just shifted from asset flop games to AI slop games. Nowadays we look at games and say "oh this is AI", but it doesn't mean you can't make a good game with AI. Obviously if we say "make the UI for my game", it will be, (from my experience), generic and not good at all. The same way we can just download an asset pack and slap it into the game. I totally agree that now we can spend time on game mechanics and quickly prototype different ideas much more easily, but it's not easy to make games. It's easy to make bad games, but still hard to make good ones.

Reply on news.ycombinator.com