A loophole on YouTube has been spotted that allows users to upload and access compressed data hidden into videos as a backup solution, although it’s not as easy or advisable as it sounds.
For it to work, Infinite-Storage-Glitch (opens in new tab) (via PC Gamer (opens in new tab)), a tool developed in Rust by Github user DvorakDwarf, must be run from a Linux distro and compiled either from source or using software development platform Docker (opens in new tab). There are no readily available releases on Github, and perhaps for good reason.
DvorakDwarf muses on the Github project page that, although YouTube’s Terms of Service (TOS) may not expressly forbid uploading videos that contain files, it is possible that Google lawyers have planned ahead, given a section about ‘circumventing the service’.
File storage in a YouTube video
As a result, although this is an interesting development, TechRadar Pro can’t expressly suggest that you rush to try this.
The tool’s creator, also being careful with their words, has warned against making this your next cloud storage solution, noting that it lacks filesystem support and comes with many bugs that they do not plan on fixing. These include bad RAM usage, which limits individual files to “about” 100MB.
Still, if you don’t particularly value your Google account, or are as interested in novel forms of storing data (like in tape and HDD hybrids, or even DNA) as us, you might find this fascinating.
Users store files inside a .zip archive, and the tool embeds them into a video file: the new one, being “several times larger”. The video can then be uploaded, as well as downloaded, with the files able to be “dislodged” using the tool.
Users have a couple of options to store their data: inside RGB pixels (quick and efficient, but more prone to corruption and sensitive to compression) or black and white binary ones.
Both modes are susceptible to compression, so several settings, such as pixel size (with 2×2 pixels recommended for binary) are available to tweak, and are stored in the first frame of a processed video to make it easier to remember options that work.
