It’s Steam Next Fest, a wonderful time of the year if you’re either a regular person who’s really into game demos or a developer whose game manages to get some shine from the Steam algorithm. With the number of participating games on a rapid, AI-powered rise, Next Fest had already, as of earlier this year, devolved into less-than-useful chaos for some. But now developers are facing a new dilemma: If AI vibecoders can copy their games based on a demo and beat them to market with slopped-up trash, is participating in Next Fest—or releasing a demo at all—worth the risk?
Some developers have already decided that, no, it’s not. Late last week, publisher Alliance Arts announced that it was cancelling a public demo of demonic Russian roulette game Drunken Goddess Reflux (which would have coincided with Next Fest) due in part to the prospect of AI vibeclones.
“We're concerned that the demo's content could be used to quickly imitate or replicate the game's presentation and gameplay with AI or other tools, and that such works could be released before our launch,” Alliance Arts wrote on Twitter, noting that it will instead be holding closed play sessions for “a limited number of participants.”
This week, Moonlighter developer Digital Sun made a similar announcement concerning its next game, ReVamp, cancelling a playtest in favor of “something more contained” on its Discord server.
“We've been concerned by recent news of games being reverse-engineered, cloned, and published before the original is even out,” Digital Sun wrote on Bluesky. “With launch still months away, a large public build of ReVamp felt like too big a risk. … Showing games early is how indies get seen and find the fans who keep us going, so we hope platforms keep strengthening protections for original work.”
“We've been concerned by recent news of games being reverse-engineered, cloned, and published before the original is even out.”
They are likely hoping to avoid the fate suffered by Needle In A Haystack, which its developers claim was the victim of code theft by a copycat, A Game About Needle In A Haystack, possibly with the assistance of gen AI. The twist is that Needle In A Haystack was already functionally a handmade copy of a still-unreleased game, Needle In A Haystack Simulator, that both benefited and suffered from a viral tweet back in August.
Steam, as it happens, is lousy with Needle In A Haystack Simulator wannabes, providing us with a handy (albeit depressing) outline of the larger problem in a nutshell. Clones of indie games that gain a measure of publicity are hardly a new thing, but AI supercharges the process. What used to take at least a few weeks or months of time and effort now takes mere days (and barely any effort at all).
With recent months bringing us a Biblical flood of vibecoded decomps and ill-considered mashups of Mario and Halo that exist not because they’re fun to play but because some hack wanted Twitter clout, many developers face a shifting landscape. Getting noticed was already a titanically tall order, especially as the number of games populating events like Next Fest began to skyrocket into the thousands upon thousands (another problem gen AI has exacerbated, though Steam’s algorithm does a better job of separating wheat from slop than many realize). But now developers who release demos run the risk of sinking into a morass of overcrowded obscurity and torpedoing future efforts to make an impression on potential buyers.
“For new devs there [used to be] two ways of making games and getting some attention,” games industry veteran Rami Ismail—whose game, Ridiculous Fishing, was infamously cloned way back in 2013—told Aftermath. “You either make a big game that takes years, build a community around it by painstakingly and consistently building community around it, and then hope that by the time you launch your Next Fest demo you have the wishlists and momentum to not just disappear into the quicksand; or you build something ‘hook-y’ fast, post [quickly], and have it ready to launch if it goes viral. If it doesn't, you just move on.”
“Both of those models are now incredibly risky,” he explained. “The slow one because sloppers can literally see what is gaining momentum and copy the aesthetic and gameplay, launch fast and simply dupe a lot of players with an inferior and cynical game. The original will remain a superior game, but the winds will be out of [its] sails, and countless people will believe they're playing the original that they've seen a few times on the internet. The faster one is risky because ultimately, the development speed and hook-y nature of the game makes the idea and basic implementation the ‘golden’ part, but they're also very fast to clone. So as soon as something [gains] momentum, anyone with resources can burn a few grand stealing it, launching it, and making sure the actual creator never sees any opportunity.”
This new status quo has left many developers in a bind.
“We're working on a small music game that's pretty far along, and we'd love to start sharing screenshots/videos/demos online,” Dan Vecchitto of Trombone Champ studio Holy Wow told Aftermath. “But the game is small enough in scope, and with a specific-enough subject, that we're pretty certain people could make AI clones pretty easily if we were to start sharing media online. So even though we probably are at the stage where we could release screenshots and videos, we're just holding onto everything until god knows when.”
"As soon as something [gains] momentum, anyone with resources can burn a few grand stealing it, launching it, and making sure the actual creator never sees any opportunity."
Would a potential clone be worse than what Vecchitto and co are handcrafting? Almost certainly. But it could still mimic the superficial appearance of their game ably enough to cause real damage.
“I'm certain the clones would be garbage—it seems like most of the stuff AI spits out is actually unplayable—but it's still annoying as hell to consider! Nobody wants to see their stuff stolen!” Vecchitto said. “In particular, our game-in-progress has a very strong color palette with a few strong (and unexpected) primary colors, as well as a few distinctive fonts, visual cues, and iconography. It feels like someone with bad intentions could easily prompt an AI with those specific style/art direction choices … which is much much more annoying than if someone were to simply rip off the gameplay mechanics but changed the style. Ripoffs have been a risk since forever, but historically ripoffs don't have anywhere near the same level of visual polish as originals because the people making the ripoffs don't have the skills, time, or interest in quality.”
Many developers believe the style of game in question matters a lot here. Simpler mechanics and art styles lend themselves more readily to imitation, whether AI or people do the dirty deed. But again, AI lowers the barrier to entry, raising the risk honest developers face simply by trying to promote their games.
“Because we’re working on a narrative adventure game, it would be difficult for someone to prompt-engineer their way through a complex branching narrative with our particular framing and style,” Ellie Zygmunt, whose studio is developing a painterly adventure game called Project Baltic Sea, told Aftermath. “That doesn’t stop someone from trying to mimic the concept and dropping a slop version on Steam. The slop version isn’t direct competition in terms of quality, but it pollutes an already thin atmosphere.”
“To manage the ripoff risk, our studio is selective about what we choose to share about the game and the platforms we post on,” Zygmunt added. “We’re not going anywhere near X, for example, because that’s like posting a sign that says ‘LOOT ME’ on your unlocked front door. We posted our new game announcement on our own website as an old-fashioned blog post. The accompanying images are largely inspiration and moodboard rather than in-game images because, again, I don’t want someone to feed our visuals into a model and reverse engineer the look. I suspect we’re not alone in this approach.”
Others are experimenting with timing relative to release date and events like Next Fest, though they fear that even if they play their cards just close enough to their chests, the end result will still hurt their games’ chances.
“We planned to go up early with a trailer and Steam page, of course, but [we’re] reconsidering now,” Martin Fieber, a senior UI dev at Remedy who’s also working on an indie game with his friends, told Aftermath. “Even though we think we have more to this than can be easily copied—story, art, etc—we still fear a potential damage if one or more copies are perceived as the game we’re making. Current sentiment now is to keep everything secret until ‘shortly’ before release, but [we] have concerns about the ability to then build a fan base/interest.”
“We have basically decided our next game will need to be a shadow drop,” Joel Auterson of Irish indie studio Bearwaves told Aftermath. “Next Fest wasn't great for us anyway—nothing of value has been lost there for a studio of our size—but not being able to develop our games in the open has been a real blow, and it's very dispiriting.”
“We have basically decided our next game will need to be a shadow drop.”
The knock-on effects of these decisions could seal a game’s fate in an industry where, increasingly, publishers choose which games to bring into the fold based on preexisting interest and community.
The sky hasn’t completely fallen yet, though, and game discovery expert Simon Carless isn’t entirely convinced that the end is nigh. In a recent edition of his newsletter, he pointed to Needle In A Haystack Simulator as an example, noting that the original has far more wishlists than all except one clone and that, ultimately, if your game hasn’t struck viral gold, there’s not a great reason for thieves to pilfer it.
“When your game is complex, already semi-popular, and is called something else—like ReVamp, or Drunken Goddess Reflux—we don’t think you should delay demos,” Carless wrote. “There’s nothing in either of these games that indicates people could search and find similar games to them and be interested. (They’re not ‘one single idea’-y—and they have a lot of personality, design decisions, and players already bought in. And trailers have already been out there, if people want to hand-copy ‘em).”
Still, perception matters, and even if an individual developer’s game is relatively safe, many are concerned about the teetering house of cards around them.
"While the conversation of 'Do we think a vibecoded clone will arise?' was not one we were particularly concerned with due to the niche nature of our game, the main concern was the shifting sentiment around [Steam Next Fest]," Birdup, a developer on football match-3 game Intergalactic Puzzle Touchdown, told Aftermath. "Many of the folks we had reached out to about demoing our game seemed to express publicly that they were spending less time engaging with the event because of how hard it was to browse the listings. To see public sentiment skewed towards 'There's too much slop and thus I am not going to bother' was disheartening as a developer, even though I cannot blame them."
Recommended
Chris Person
