AI Cloners Are Here. Are Indie Devs Cooked?

Thank you so much to everyone who has reached out to ask if I read about the Haystack Games Controversy (here is a link to one article) and then Drunken Goddess Reflux (here is a link to an article about it). You are probably here because I sent you a link after you asked me “Are we cooked?”

No we are not cooked.

Wait what are you talking about?

In short, a developer named Nas tweeted a video of a high level prototype of their game about finding a single needle in a stack of 5 million pieces of hay.

The post went viral and within a week of that viral post, the developer had a Steam page live for Needle In A Haystack Simulator. Unfortunately some other developers also created Steam pages for clone versions of that game. One of the clones actually got a demo out and another released a full version of the game before the original developer could.

If you have spent any time in game dev Discords, YouTube channels, or sub-reddits, this AI cloning thing has absolutely spooked game devs. It is a controversy that combines AI AND someone stealing your game idea AND over-saturation! It is like someone in a lab took every indie anxiety and distilled it into a serum and injected into a developer’s brain stem the moment they installed Godot 4.7.2.

There is a lot of general anxiety and this is complicated so I will try to disaggregate whether we are cooked or not:

TL;DR

Is my game going to get cloned? – No, your game is not going to get cloned and then turned against you unless it meets ALL of these criteria: a non-bespoke art style, a fantasy that has gone EXTREMELY viral already, represents a genre that has already been explored by other games.
Should I even market my game because then AI Bros will steal it? – Yes you should still market your game, and you should still do it nice and early.
Should I just hide in a game dev closet and shadow drop my game so AI Bros won’t find me? – No. But if you suspect that your game is based on a quick fun “gimmick” instead of a deep experience, you should just be ready to react with a Steam page and a demo close behind just in case. But 99% of games are not this.
Should I make a demo because AI Bros will decompile it and cook me? Yes you should make a demo, they are still the best way to increase visibility on your game.
Should I shadow drop my game – No. This is worse than being cloned by an “AI Bro”
Are indie games cooked? – No, indie games are not “cooked.”

The absolute anxiety cloud hovering over the indie game industrial complex has you all looking like a bunch of sad sacks. We are video game designers here which means (other than theme park designers, candy makers, and liquor salesmen) we should be the most fun people in the world. So let me start this blog with an indie pep talk to lift up the spirits and then I will get into the data and analysis.

🎉Indie pep talk🎉

Anytime something gets really chaotic and complicated with everyone worrying about AI, crypto, saturation, bananas (etc), I find it important to do a Zack Morris Timeout and repeat my first principle about this industry:

People fucking love video games to a degree that is kinda scary. Steam shoppers (specifically) spend so much money on video games because they love them, they obsess over them, spend every waking minute thinking about them, arguing over them, and actually paying for and playing them. Steam shoppers like real video games. This is their hobby and passion. Making and becoming successful in indie game development is difficult but it is not impossible because these sickos exist.

The first principle of video game development (Also, I call them sickos lovingly.)

Sometimes I wonder if I am being overly optimistic or ignoring something when I write these “golden age” and “don’t fear” posts. But, when I step back I have to remind myself that we are video game developers and we are working in the #1 creative medium of our era.

I recently saw this chart and read this blog that compared how people spend their time in 1995 vs 2026.

Look at us. “Playing a home video game” sticks out like crazy. We are winning! By a lot. Think about that!

It may be hard but at least you aren’t trying to:

Make movies (down 8%)
Write books (down 18%)
Run a sports team (down 6%)
Run a bar (down 3%)

I listen to podcasts and read blogs written by indie authors, film-makers, and musicians and often I hear them lament how their creative field is slowly dying. They frequently ask “how do we make our work as exciting as video games?” They blame us video game developers for the decline of films, books, and music.

We are winning. We are cooking (and eating) all other creative mediums.

People love, ABSOLUTELY LOVE, playing video games. The Steam playerbase is full of total sickos who buy games they will never play. The platform of Steam itself is growing every day. For the first time ever they hit 43 million concurrent players on October 11th.

Steam players like our games so much they hack them to add new features. Kids spend so much time on what we make that their parents have to punish them to stop. Every backpack I see at my son’s school is Pokemon, Minecraft, Mario, or some weird character that I assume is from Roblox. Most of my kid’s friends are dressing up as video game characters for Halloween. The top Hollywood movies are based on video game IP like Minecraft (a former indie game) and Super Mario. Even indie horror game-turned-movie Iron Lung out-performed expectations.

We are winning. We have cultural dominance. There is no other creative field where it is more possible for a small team to make a living directly from what they create. Can you imagine how hard it is to be a full-time writer when there are literally 250,000(!) books released every month on Amazon and readership is declining because we are becoming a post-literate society?!? Indie authors laugh at how good we have it.

Yet we act like we are losers and the world is ending.

We are winning so hard that, honestly, that Data for Progress chart has me a bit sad. I am a little worried games are a bit too popular. I want us all to be able to make a living doing this but … a 22% increase?!? While at the same time there is a decline in offline, tactile, in-person, human contact.

Could we actually be the cigarette makers right now in this human-social health crisis?

Are we the baddies?

I mean for the love of humanity, please stop doom scrolling Reddit and Youtube, maybe spend a little less time playing games, and then go read a book while on a picnic, and then ask your neighbor if they would like to come over on Halloween night and have some ghost-themed cupcakes you made (that counts as entertaining people at your home BTW!)

Wait this was supposed to be a pep talk.

Games are cool, at least go to a movie at a theater.

That will uplift you.

Boom!

🎉Pepped!🎉

Let’s go.

Ironically “AI bros” are scared too

All day every day I get emails and Discord DMs from developers asking “am I cooked because someone will make an AI clone of it?” But also a portion of my audience asks me “Am I cooked because I am making a game with some AI art in it. Will people yell at me and will it hurt my sales if my game is considered AI slop?”

People like to blame “AI Bros” who tweet “I one-shotted GTA clone with one Fable 5 prompt. Game devs are cooked.” But for the most part, the people posting those “one shot” tweets are bros selling a product / service. But, the people making most of the “AI slop” you see in Steam Next Fest are just hobbyists who don’t yet have better game dev skills or people living in developing countries who can’t afford to hire an artist so they make do with AI. They are just not that good at making games yet and so their work looks crude. This is not to excuse AI cloning. Do not clone games. But, I just think the enemy in the AI debate is too simply depicted as an “AI-bro.”

So I should say hobbyist devs who use AI to augment their skills (and may or may not be cloning or doing a fast follow) are scared too.

What actually happened?

I think I have read just about every article, Reddit thread, and YouTube video about the “Haystack” situation and it seems to always come down to someone saying “Haystack…. AI Slop Games…. Am-I-Rite?” and then the comment section just erupts in a doom loop about indie games while simultaneously launching into a screed against AI.

Everyone is worked up into a tizzy but nobody ever took the time to look day by day at what happened.

Take a breath. Time Out. Let’s disaggregate.

The data comes from SteamDB for a selection of Haystack games.

Needle-in-a-Haystack Games: Steam Followers

Aug 22 – Oct 28, 2026 · Hover the numbered badges for what happened

Let me give you some more context for this timeline and those callouts. Each callout you see pertains to this list:

Original tweet goes viral.
Steam coming-soon pages go live for the original and two clones on the same day. The original’s developer starts running Tiktok ads but under a different name of Studio BitDot. Original paid Tiktok. Also on August 29th there was coverage by more content creators.
Needle In A Haystack Simulator (Original) playtest starts. 23,000 people sign up. Notice the playtest is actually when the game gets its first wishlist spike (533 followers). The original game is way ahead of all the other clones at this point and it goes up and up from there. A playable version of the game is one of the best things you can do.
Needle In A Haystack Simulator (Original) gets a publisher with Polden Publishing (well the publically set their publisher to that on that date.)
The clone game Find The Needle launches a public demo (not just a playtest.) This is the first easily playable version of any of the Haystack games.
It takes 13 days for the Find the Needle demo to slowly bubble through YouTube content creators. On 9/23 Orbital Potato finds the demo and you can see the velocity at which the clone Find The Needle accelerates and actually surpasses the original game in follower count on October 5th.
The first full release of a Haystack clone drops on the store for sale. It has nearly the same name as the original except it is lacking the word “Simulator.” From here on out I will refer to it as Needle In A Haystack (App ID 5085740). Note however that this game launches with a very low follower count, and its release does not slow down the momentum of the two front runners of Find The Needle and Needle In A Haystack Simulator.
Clone Find The Needle takes the demo to the October Steam Next Fest. It does great and is one of the top demos of the fest.

Indies are learning the wrong lessons

Please reread that 2 month marketing breakdown and the follower chart above because it is much more complicated than just “AI Clones take over everything and we are cooked.”

I have noticed that many of these scared indies are reacting and making decisions that run counter to what actually happened.

Here are the most common incorrect reactions:

“If I get cloned, I am dead, game over, I am cooked” – Not really. The original Needle In A Haystack Simulator is still growing at a healthy 60-70 followers / day. That is about 600+ wishlists / day (and that is despite not having a demo live. They should get one out there!) I estimate that the original game has about 20,000 fewer wishlists than the lead clone. That sucks but, all things considered, not that bad since they still haven’t done Steam Next Fest and they don’t have a demo. They also signed with Polden Publishing on September 1st. So they aren’t dead in the water.

And look at this period of the chart when the clone released the first demo:

Notice how the demo and content creator coverage was also increasing the velocity of the original game too!

Also if your game does get cloned you get to reach out to the press and say “my game got cloned!” The press has a known negativity bias and they looooove to write stories about something bad happening to someone. You can ride that negativity bias to the bank. You also get to do a post on Reddit and YouTube complaining about it. So soak in the negativity (then turn it off so you can get back to work.)

“I will just keep everything secret and shadow drop” – This is a sign of someone falling for the loss aversion bias. Please do not shadow drop your game! The cloner who made Needle In A Haystack (app ID 5085740) tried to be first to market and released their game 1 month after the viral tweet. But they just did a Leroy Jenkins.

They only had 452 followers at launch (estimated about 4,000 wishlists) and as you can see their following flatlined while the main 2 haystack games took off. Even AI-cloners can’t shadow drop a game and have it work. You still need to collect a lot of wishlists, do a demo, do Steam Next Fest. 6 months minimum. If you shadow drop you miss out on all the features you are paying for with your 30% Steam cut: New & Trending Free, Steam Next Fest, Personal Calendar, (hopefully) New & Trending and Popular Upcoming. Don’t surrender those because you are scared.

“I am not going to make a demo because cloners will steal my game, I will just run a closed playtest” – This is a sign the developers didn’t observe what actually happened with Haystack Sim! This playtest-only approach is exactly what Needle in A Haystack Simulator did and they were still cloned and their popularity was passed when “Find The Needle” released a demo first. YouTubers like playing demos more than playtests. You must be first to market. Be brave. Don’t be a nit. Get that demo out there.

“There are so many clones nobody can find any of them” – Again look at the follower chart. There really are only 3 haystack games that got more than 100 followers. Steam is good at promoting games that already have momentum and really good at hiding games with none. The first mover advantage compounds. Just try it out to see if you can find them: go to Steam and search the word “Haystack.” The number 1 game is the original. Type “needle” and the top 3 results are the top 3 games in my chart. If you go to Google and type “needle game” you would expect some totally vile stuff but NOPE, the original and the #1 clone are both at the top of the search results. The dozens of follow-on clones are all buried. It isn’t that hard to find the top contenders.

This happened with the “Banana” game trend of 2024. Back then there were also so many cloned games of clickable (mostly edible) objects. But, only the first game Banana did well.

“If I release a demo the cloners can just decompile it” – Steam has a bias toward first mover advantage (with quality). With Haystack the first game and the 2 clones were the winners. The first game with a demo was the winner. The Needle In A Haystack clone that released on 9/29 didn’t do well. It wasn’t good enough. But that didn’t hurt the top ones. Your goal is to be the first to release a quality demo and a quality game. That hasn’t changed.

“AI bros just throw out slop as fast as possible and Steam players will eat it up like piggies” – Go back and read my first principle of game development. Clone #3 Needle In a Haystack (App ID 5085740) was the first to market. But it missed the mark. Something about it just isn’t as fun and it stalled out while the other 2 haystack games are still rising. I played Find The Needle (you should too in the interest of science and understanding our cloning future) and I have to admit…. it isn’t that bad. There is definitely sloppiness (there are constantly like 2 songs playing and a lot of warnings about low frame rate) but the core loop is good and interesting. YouTubers played it a lot. They are very discerning about what is fun, and they liked it. I am not condoning the cloning that Find The Needle did, but I can’t lie, the game is fun.

“Marketing is cooked! We gotta throw out the playbook” – Not really. The top Haystack game used the standard indie marketing playbook. Get a Steam page up, get to a demo as fast as you can well before Steam Next Fest, reach out to content creators to get momentum, do Steam Next Fest. That’s the playbook and it worked great (for the cloner, and pretty good for the victim).

“AAAAAAAH!” – I think you are just stressed. Step back. Close YouTube. Close Reddit. Anytime someone posts “my game got cloned” news, look up the main game then the clone on SteamDB. Usually the clone has a fraction of the follower count and if they try to do a leroy-jenkins marketing launch they always fail.

What has changed?

I do think getting a demo out is the best plan and that has not changed and is the cornerstone of doing well on Steam. Needle In A Haystack Simulator was in the lead until the copycat put out a demo and content creators played it. You need a demo.

Now the thing that has changed is maybe the speed you have to do this. I do recommend a playtest before a demo, but if you have a proven viral game, and the clones are popping up, go straight to demo. You need to make sure you are the first game that the content creators play.

I also think some games are more susceptible to being cloned than others and that will change the timing of when you talk about your game and how.

In my next blog I will be looking at how to determine if your game might be cloned and whether you should hold back on marketing it.

Worse-case scenario

I really feel bad for the original Needle In A Haystack Simulator. This is bad. But for all other indies take heart. This is like a perfect storm of events that aligned on one game:

A really funny viral idea that is one of those that make you slap your head and say “why didn’t I think about that?” This is very rare. Usually I see only one or so of these a year. Past ones are Banana, Vampire Survivors, Choo Choo Charles. Unfortunately Haystack is a legendary viral idea.
Simultaneously, the idea is easy to clone because it follows core genre conventions of a very popular genre called “clean up” games. Look at this comment under the original viral tweet that accurately predicted how the game will be played:

Everyone knew the structure of this game after reading that initial tweet. That isn’t to say the game is bad, it’s just that the game design almost wrote itself. That is much easier to vibe-clone.

One of the clones is actually competent. Find The Needle isn’t a shining example of fine craftsmanship, but it has it where it counts. And the developers didn’t just “one shot” it and slap it on Steam. The developers recently posted that they are on “version 8” of the demo. They are actually working on the game based on feedback like any other developer should.
Needle In The Haystack Simulator (Original) is the first game to be exposed to a novel AI clone attack. At the time, we had no known natural defenses. Now we know. If you might have a viral game on your hands, proceed quietly, and get that demo ready. Move quickly but cleanly.

Do not freak out. Do not abandon game development. This is bad, but it will likely not happen to you.

That fear you are feeling is called the availability heuristic. It is this mental defect we humans have where we think something is more likely to happen based on how likely it is to come to mind. So a vivid story about the theft of a developer’s game is front of mind and every time you open Reddit there is someone doom posting about it. So it is front of mind for all of us. That doesn’t make it likely to happen.

It is like an airplane crash. Horrific, yes, but you are more likely to be injured on the way to the airport than on the flight itself.

Stupid human brain. Once you can identify the availability heuristic, that should make you feel better.

Time out. Take a minute. Identify it. Get off Reddit. Get back to making your game.

Feel better now?