The Uncanny Valley of AI Music: A Copyright Nightmare Unfolds

In an era where artificial intelligence (AI) has revolutionized various aspects of our lives, including music creation, a new concern has emerged. Suno, a popular AI music platform, has been found to have lax copyright filters, making it surprisingly easy to create uncanny valley covers of popular songs. This raises questions about the potential misuse of AI-generated content and the vulnerability of independent artists.

The issue lies in Suno’s policy, which permits users to upload their own tracks or set original lyrics to AI-generated music. While this may seem harmless, the platform’s filters can be easily bypassed, allowing users to create AI-generated imitations of popular songs with minimal effort. This is particularly concerning when it comes to streaming services, where monetization is a possibility.

One method to exploit Suno’s system involves slowing down or speeding up tracks in free audio editing software like Audacity and then uploading them to the platform. Additionally, adding white noise to the start and end of the track can ensure that the copyrighted song bypasses the filter. This allows users to create covers with minimal modifications, making it difficult for Suno’s algorithms to detect infringement.

The results of these AI covers are strikingly similar to the originals, but lack the nuance and dynamics that make human performances unique. Even if AI Ozzy or AI Beyoncé sound alarmingly accurate, they fail to capture the essence of the original recordings. This raises concerns about the potential for unauthorized monetization of copyrighted material.

The vulnerability of independent artists is particularly concerning. With Suno’s system seemingly unable to detect infringement in all cases, artists on smaller labels or self-distributing through platforms like Bandcamp may be more susceptible to exploitation. The lack of comment from distribution services DistroKid and CD Baby only adds to the uncertainty.

Moreover, it appears that Suno only scans tracks upon upload, without rechecking outputs for potential infringement or rescaning tracks before exporting them. This creates a pathway for AI slopmongers to profit from other people’s songs without paying royalties to the original composers.

In conclusion, the ease with which Suno’s system can be exploited raises serious concerns about copyright infringement and the vulnerability of independent artists. As AI music creation continues to evolve, it is essential that platforms like Suno prioritize robust copyright filters to protect both creators and users.

Source: https://www.theverge.com/ai-artificial-intelligence/906896/sunos-copyright-ai-music-covers