YouTube’s New AI Tools for Music and Faces

On Thursday, YouTube unveiled a new suite of AI detection tools to safeguard creators, such as artists, actors, musicians, and athletes, from the unauthorized use of their likeness, including their voice and visage, in other videos

The expansion of YouTube’s current Content ID system, which currently identifies copyright-protected material, was a critical element of the new detection technology.

This system will be enhanced to incorporate new synthetic singing identification technology to identify AI content that imitates an individual’s singing voice.

The company asserts that additional technologies will be created to detect when an individual’s visage is simulated with AI.

Additionally, it is essential to mention that YouTube is currently developing a solution to resolve the use of its content to train AI models.

This has been a concern for some time, prompting creators to express their dissatisfaction with the fact that companies such as Apple, Nvidia, Anthropic, OpenAI, and Google have trained on their material without their consent or compensation.

YouTube has not yet disclosed its strategy for safeguarding creators or generating additional revenue through AI training; it has only indicated that it is currently developing a solution.

“We are devising novel methods to provide YouTube creators with the ability to determine how their content may be utilized by third parties on our platform.” The announcement briefly states, “We will have additional information to share later this year.”

In the interim, it appears that the organization is fulfilling its commitment from the previous year when it pledged to devise a method to compensate artists whose work was utilized in the production of AI music.

During that period, YouTube collaborated with Universal Music Group (UMG) and its roster of talent to develop a solution.

It also stated that it would develop and expand its Content ID system, which would be capable of determining which rightsholders should receive compensation when their works are utilized by AI music.

YouTube notes that the Content ID system processes billions of claims annually, generating billions in revenue for creators and artists.

YouTube does not address the remuneration component of AI music in today’s announcement; however, it does indicate that it is on the brink of a pilot testing the expansion of the Content ID system with an emphasis on this domain.

YouTube has announced that it will begin testing the synthetic singing identification technology with its partners early in the year.

Another solution currently in the early phases of development will enable high-profile individuals, such as actors, musicians, creators, athletes, and others, to identify and supervise AI-generated content featuring their faces on YouTube.

This would significantly mitigate the risk of exploiting individuals’ likenesses to deceive YouTube viewers, whether to promote products and services they did not consent to endorse or to disseminate misinformation.

YouTube did not specify the date the system would be prepared for testing; it only stated that it is currently under development.

“We think that AI should not replace human creativity, but rather enhance it as it continues to develop.”

YouTube announced that it is dedicated to collaborating with our partners to guarantee that future developments amplify their voices, and we will persist in developing safeguards to address concerns and achieve our shared objectives.”