Let me be upfront about something before we get into it: AI is one of those topics where I think nuance tends to get flattened fast. I completely understand why a lot of creatives — artists, musicians, writers, developers — have strong and formal objections to AI tools, particularly around generated art and audio. Those concerns are legitimate and worth taking seriously. I have my own opinions about where the lines are, but I don’t see this as a black and white issue. As a tool that assists rather than replaces, AI has genuine potential. As a tool that’s handed to someone who uses it to skip the parts of the process that would have taught them something or is pushed forward as a decision instead of an iteration, it’s a different story.
It’s also worth noting, briefly, that “AI” has become one of the most liberally applied labels in software right now. A lot of tools that would have just been called “smart features” or “automation/procedural” three years ago are now marketed as AI. That’s a whole separate conversation — and probably a future post — but it’s worth keeping in mind that not everything wearing the label is the same thing.
One more thing on the team and tools angle: if you’re working with other people on anything creative, it’s worth making sure everyone is on the same page about which tools are in play. That was true before AI and it’s especially true now. A collaborator who doesn’t know you’re using a particular tool is a collaborator who can’t make an informed decision about what they’re contributing to.
What Is A Game Jam
For the uninitiated: a game jam is a timed event where participants — solo developers or teams — build a game from scratch within a set window, usually around a theme announced at the start. They range from a few hours to a few weeks, but the format that tends to define the experience is the 48 hour jam. Two days. A theme you’ve never seen before. Whatever you can build, test, and submit before the clock runs out.
Game jams have been around for decades but they’ve changed dramatically even in the last ten years. Discord made remote collaboration seamless. COVID forced the entire concept online and accelerated a shift that was already happening — from shared physical spaces to shared digital ones. What used to mean a room full of people, a pile of energy drinks, and someone’s laptop hooked up to a projector now looks like a dozen people in different time zones, a shared repository, and a voice channel that stays open for 48 hours (if you’re lucky). The communal spirit survived the transition, mostly. But something about the texture changed. That evolution is worth its own post — but it’s important context for what comes next.
AI Walks Into A Game Jam
Here’s where things get interesting. A 48 hour game jam is, by design, a quick and dirty exercise. The goal isn’t a polished, shippable product — it’s a prototype that demonstrates an idea, proves a concept, and exists as something playable by the end of the weekend. Speed is the whole point. Scope is the enemy.
AI tools — particularly in coding and game engine workflows — are very good at quick and dirty. The kind of boilerplate code that used to eat hours can now be generated in minutes. Systems that required a developer to know exactly what they were doing can now be scaffolded by someone who knows roughly what they want and can iterate from there. In a 48 hour jam, that compression of time is significant. What used to be a sprint might start to feel like a jog, which means more time for the parts that actually make a game feel like a game — the polish, the feel, the details that separate something forgettable from something people remember.
There’s also a more basic impact: more submissions. One of the quiet tragedies of game jams is the number of teams who run out of time and submit nothing, or submit something so incomplete it barely counts. AI as a development accelerator could meaningfully reduce that — giving teams enough runway to actually finish, or at least finish enough. That’s not nothing.
What Game Jams Were Actually For
But here’s the thing that gets lost in the efficiency argument.
Game jams were never just creative races against the clock. They were communal. They were designed around the idea that making something — even something small, even something rough — is better done with other people. Not just because more hands means more output, but because the process of figuring out how to work with other people under pressure, with a deadline, toward a shared goal, is itself the lesson. You learn things in a 48 hour jam that you can’t learn any other way. You learn how to communicate quickly. You learn how to make decisions when there’s no time to second-guess them. You learn what you’re actually good at when everything else falls away. And you learn, maybe most importantly, that the thing you make together is always better than the thing you would have made alone.
AI doesn’t erase any of that. A team using AI tools is still a team. The collaboration can still happen, the learning can still happen, the communal pressure-cooker experience can still happen. But it doesn’t enforce it either. And that’s where I think the more interesting concern lives.
The Village Is Optional Now
The game industry has a tendency toward what I’d call toxic independence — the idea, particularly common among newer developers, that the goal is to do everything yourself. To be the one-person studio. To need no one. It’s understandable where it comes from. There’s a certain mythology around the solo developer who built something extraordinary alone, and tools that make solo development more achievable feed directly into that mythology. As well as a healthy dose of thoughts like, “no one wants to help me make my game,” when they try to pitch an idea to others.
AI accelerates this. If you can generate your code, generate your assets, generate a soundtrack, and build something submittable in 48 hours without ever collaborating with another human being — why would you? The jam becomes a solo exercise. The village becomes optional.
And that would be a loss, not because collaboration is always comfortable or efficient, but because the game industry is not those things, it is not a solo medium. Almost nothing worth making gets made alone. The skills that game jams were designed to build — how to communicate, how to divide work, how to trust someone else’s judgment, how to make something together that neither of you could have made separately, how to work on something you might not have a terrible amount of interest in — those skills don’t develop in a room with one person and a very fast AI.
My concern isn’t that AI takes over game jams. My concern is that it makes people think they don’t need to show up to the village, right at the moment in their development when showing up to the village is exactly what they need.
The fastest team in the room might not be a team at all. And that’s worth paying attention to.