# Side Projects in the AI Era: The Ponzi Potato Story

In 2011 I wrote my bachelor thesis about **why video games matter**. In 2014 I sketched a game called *Save Me*. It never left the sketchbook. Since then I've built secret sharing tools, file transfer tools, websites and way too many side projects - but never an actual game.

Fifteen years later, it finally happened. It's about potatoes. Of course it is.

## **Blame Universal Paperclips**

If you've never played [Universal Paperclips](https://www.decisionproblem.com/paperclips/): don't. Or do, but clear your afternoon first. You start by clicking a button to make a single paperclip. A few hours later you're turning the entire universe into paperclips. It's mostly numbers and buttons, and it's one of the best games I know.

What got me is how much story you can tell with almost nothing. No graphics, no characters - just a number that keeps going up, and a growing feeling that maybe this isn't a great idea.

I always wanted to build something like that. I just needed something to replace the paperclips.

## **Plant a potato, build an empire**

[Ponzi Potato](https://www.ponzipotato.com/) starts with one button: **Plant potato**. It grows for 12 seconds and turns into 2-5 potatoes. Plant those. Harvest. Repeat.

Soon you're running a business. Then come the machines, more land, fries, chips and vodka (obviously). Then bioplastic, potato power, potato futures and a quantum computer. In the end, the universe is one big potato field. Sorry, universe.

And then there's the pyramid. Every incremental game is a bit of a pyramid scheme anyway: you invest now so the numbers go up later. So why not make it official? Invite friends, they invite friends, everybody "earns". There's a Ponzi fund called **Spud Capital** that takes all your cash. You can even *rent* friends on the black market if you don't have enough real ones.

The best part: the pyramid only ever pays in potatoes. Nothing on the site is an investment. Nobody gets hurt, except maybe your productivity.

## **Just an experiment**

Full disclosure: there's no business plan. Ponzi Potato is an experiment, and the main goal was to learn something new.

Building a game is surprisingly different from building a product. In a product, you try to make things easy. In a game, you make things *just hard enough*. How long should the first upgrade take? When does waiting get boring? How fast can a number grow before it stops meaning anything? I spent more time tweaking timers and prices than writing features. I even ended up building a little simulator, just to see if a player would get stuck somewhere around the vodka stage.

Also new for me: keeping cheaters out. The game runs in your browser, but the server replays everything you do and has the final word. Editing your local save does nothing. Sorry.

## **The AI did the heavy lifting**

Here's the honest part: I didn't write most of the code. AI did.

I was the game designer, the product owner and the annoying tester. I said things like "the golden potato should feel more golden" and "the press cameras need to flash all at once". The AI wrote the code, I played, complained, and we went again.

The whole thing took about a week, and around 180 commits. A few years ago, a game like this would have been a "maybe one day" project for me. Probably right next to *Save Me*.

Does that make it less mine? I don't think so. The ideas, the bad jokes and the potato puns are all mine. The AI just did the typing (a lot of it).

## **Because I can**

Honestly, that's the real reason. Because I can, and because it's fun.

Side projects are play. Nobody asked for a potato pyramid scheme. There are no deadlines and no stakeholders - just me, a stupid idea and a free weekend. Or seven.

## **What's next?**

No idea. Maybe more stages, maybe a mobile version, maybe nothing. Right now I'm just happy it exists.

Go plant a potato at [ponzipotato.com](http://ponzipotato.com) and tell me how far you get. I'd be happy for ideas, feedback and bug reports. Or just recruit me into your pyramid :)

Best, Chris