Steal the Beat

by | Nov 8, 2023 | Artwork #3: Intervene, Projects

All players start with the same 30 second music video clip. The first one to upload it to YouTube without their copyright detection recognizing the song wins. They may use any means necessary to do so – distorting the video, distorting the audio, rerecording parts of it, renaming the upload, etc. Anything is fair game as long as the clip stays recognizable.
You will likely need a 3rd party judge to determine if the video clips are intact enough to be valid.
When you are done, share your strategies in hushed whispers where YouTube and copyright holders cannot hear.

To playtest this game, I played it single-player with Mariah Carey’s “All I Want For Christmas Is You”. You can see my results in the video below.

Each attempt took about 10 minutes to render and export, and YouTube recognized the copyrighted material in attempts 1 and 2 in less than 20 seconds. Whether or not my third attempt counts as “intact” is up to your interpretation.


Artist’s Statement:

This game ended up significantly harder than I expected. To start, you can only even play it if you have the software and knowledge to edit video and audio sufficiently. I used Adobe Premiere in my playtest, but it could be possible with entirely free software as well. Once you actually do find some players and start, the game is very slow. Waiting to render and export every attempt becomes tedious quickly. As you can see in the video, I had to distort the audio and obscure the video very severely to be able to upload it. The game may be easier if you play with a less popular song, where the copyright is enforced less harshly.

I was inspired by Jennifer Gradecki and Derek Curry’s intervention in algorithms and computer systems. Their artwork is very concerned with algorithms and data processing, and how they are applied and misapplied in our modern era. I am fascinated with copyright, and have seen various creators on YouTube employ different strategies to play short relevant clips of copyrighted material in their videos. I figured this copyright dodging could make an interesting creative game. There are many ways to go about avoiding YouTube’s copyright detection, but in practice you realize how difficult a task it can truly be.

Part of what I wanted to do with this was to teach players how to get around YouTube’s system, and help them firsthand understand what makes YouTube stop recognizing the copyrighted material. This desire to teach is also why I urge players to spread the information they learned, without telling copyright holders. I think the current systems of copyright and its enforcement are detrimental to art, and the allocations of fair use are insufficient. I wanted to subvert this system and intervene in it. By playing the game on YouTube, players directly interface with perhaps the most-used copyright enforcement algorithm on the entire internet. One thing the game does very well is show how difficult it is to subvert YouTube’s automatic detection. When copyright laws were written, such algorithms did not exist, and copyright infringement had to be enforced on a case-by-case basis. This was slower, but allowed more edge cases to slip through the cracks. Nowadays, there is no way to slip through unnoticed when YouTube checks every video for copyrighted content. Sometimes the copyright holder chooses not just to take any monetization on the video, but to block it entirely. Jennifer Gradecki and Derek Curry’s Boogaloo Bias artwork (among several of their other works) is very concerned with false positives in automatic detection algorithms. This is a huge issue in YouTube’s copyright detection system that I wish my game did more to address. Perhaps one could play to try and twist one song to be recognized as another, but that is a game for another day.