Benn Jordan has made an algorithm that can detect if music has been made by AI or notBenn Jordan has managed to create an algorithm that can detect when music has been made with AI technology.
In a new YouTube video, Jordan unpacks how using the power of missing song data from file compression can catch and identify artists who download AI-generated music from sites such as Suno and Udio, and then upload it to streaming platforms such as Spotify to make royalties.

READ MORE: Kits.AI removes ad encouraging producers to train custom AI vocal models using Splice samples — here’s why that matters

So far, Jordan’s AI model has been able to detect if a song has been generated by AI on Suno with 100 per cent accuracy. For the experiment, he scraped 560 of Suno’s top generated songs and staff picks, and alleges that out of those 560 only 11 were not being monetised by posing as a human artist on a digital streaming platform.
To further explain, Jordan unpicks how most humans cannot tell the difference between compressed and lossless audio: “Ultimately our brains are just collecting pressure waves and then passing them through latent inhibition and then only really registering data that’s useful to us, and audiovisual file compression has managed to make a surgical art out of exploiting that distinctly human information processing workflow,” he states in the video.
“Earlier this year it dawned on me that very little of what generative AI companies were non-consensually scraping off the internet was uncompressed or lossless, they scraped things like YouTube and Spotify and SoundCloud to make their data sets generate AI music.
“A neural network of this type – even if it’s a convolutional neural network – is a black box, you can’t tell it to note these limitations in a functional way. Even if you could, you can’t technically describe the differences you might hear in compressed files, you just say ‘it’s glitchy sounding’, or ‘it sounds a little bit off’, and that’s useless information in machine learning.”
Jordan continues, “My point here is that 75-to-90 per cent of information that humans don’t realise is missing when listening to compressed audio could potentially be used as a fingerprint when identifying if something was created with generative AI, and potentially even identifying the content that was scraped in the data set.”
Benn Jordan has revealed to MusicTech that he plans to speak to TuneCore, DistroKid and other distribution platforms “about changing their terms to disallow 100 per cent AI-generated content, and revoke royalties to people uploading Sudo/Udio creations while posing as human artists.” He goes on: “I don’t care if there’s AI music on Spotify. I just care it’s siphoning royalties away from musicians.”
You can watch the full video below to find out more:

The post Benn Jordan has made an algorithm that can detect if music has been made by AI or not appeared first on MusicTech.