YouTube Sets Parameters for AI-Generated Music Content

YouTube, owned by Alphabet, stated that it’s currently working on a mechanism that will enable music partners to request the removal of content that replicates the unique vocal qualities of artists.

Initially, this novel takedown request system will be accessible to labels or distributors representing artists involved in YouTube’s AI music experiments. However, YouTube envisions expanding its availability to other labels and music distributors over the upcoming months.

In August, YouTube entered into a collaborative agreement with Universal Music Group (UMG) to co-develop AI music tools envisioned to provide protection while creating opportunities for music partners. As a part of the initiative, YouTube established an “AI Music Incubator” to foster the creation of new AI tools by enabling artists to collaborate with YouTube developers.

Among the notable figures involved in the incubator are Brazilian sensation Anitta (recently signed to Republic Records under UMG), ABBA’s Björn Ulvaeus, hitmaker Louis Bell, budding artist d4vd, the estate of Frank Sinatra, and neo-classical composer Max Richter, to name a few.

It’s important to note, however, that takedown requests submitted through YouTube’s new mechanism won’t be granted automatically.

“In determining whether to grant a removal request, we’ll consider factors such as whether content is the subject of news reporting, analysis or critique of the synthetic vocals.”

YouTube Product Management Vice Presidents Jennifer Flannery O’Connor and Emily Moxley via YouTube

Furthermore, Google, YouTube’s parent company, has reportedly been engaging in discussions with UMG and Warner Music Group (WMG) to develop a solution that allows rights owners to “collect payment on revenue generated by fan-created AI deep fakes.”

YouTube has plans to introduce a label within the video description section to inform viewers when the content is created or generated by artificial intelligence or is synthetic in nature. For videos concerning “sensitive topics,” this label will be directly displayed on the video panel. This approach seems to echo TikTok’s policy announced in September, in which the platform encouraged creators to add a label to AI-created content uploaded to their sites.

In a noteworthy omission, YouTube did not clarify in its blog post if it is developing technology capable of autonomously recognizing AI-generated content, a claim that TikTok has made openly. In a related development, the music streaming service Believe, during an earnings call last month, announced the creation of AI-detection tools that can identify a deep fake audio file with an impressive accuracy of 93%.