The Generative AI revolution has been sweeping through various industries, and music is no exception.
Many AI music tools now quickly generate original, studio-quality songs from a simple text or button prompt.
While this groundbreaking development improves creativity and the music production process, there are many concerns and potential implications that must be considered.
At SOUNDRAW, we’re well aware of AI's potential impact and are committed to maintaining ethical AI in music.
In this article, we’ll address some of these concerns and also share our approach to AI music.
The Rise of AI in the Music Industry
In recent years, the rise of AI music creation has been due to rapid advancements in cloud computing, machine learning, and deep learning.
AI models such as OpenAI’s GPT-4 model have been trained on large datasets of existing music to understand patterns and styles, so they can generate original compositions with little user input.
Modern AI tools can now generate high-quality music across various genres. It’s not just music content; AI is also used to write lyrics, create artwork, and master songs.
Most artists and music producers use AI to augment creativity, automate some production tasks, and gain new inspiration. Creators also use AI to generate theme-based music for their creative works.
This transformative shift in music production opens several innovative possibilities. It also raises ethical concerns which includes the use of copyrighted music works in training AI algorithms.
As the music industry continues to embrace AI, addressing these concerns is more important to ensure a fair future for music creation.
Ethical Concerns with AI Training on Copyrighted Music
In 2023, a producer named Ghostwriter made an AI-synthesized song titled “Heart on my Sleeve” and released it on streaming platforms.
The song was an instant hit because it mimicked the voice of famous pop artists Drake and The Weekend and sounded realistic. It went viral and took the internet by storm, allowing the music world to witness the insane capabilities of AI music.
In response, Universal Music Group, which represents Drake and The Weekend, requested that the song be removed from Spotify and Apple Music. The music industry giant also asked the streaming platforms to block AI companies from analyzing melodies and lyrics from copyrighted songs to create new AI-generated song pieces.
It’s not just this singular case; looking around the internet, you will find songs sung by AI versions of real artists. This occurrence raised severe concerns about using generative AI to infringe on copyright music and create deepfake sounds that mimics real artists.
The rise of AI in music creation has elicited mixed reactions from artists and composers. Many see it as a powerful tool for inspiration and experimentation, while others express concerns about its impact on their profession and the essence of musical creativity.
Grammy-winning producer Oak Felder, known for his work with Demi Lovato and Alicia Keys, views AI as a potential collaborator rather than a threat. "AI could be an incredible tool for creators," he states. "It could help with writer's block or generate new ideas we hadn't considered."
On the other hand, Nick Cave, the frontman of Nick Cave and the Bad Seeds, has been more critical. In response to AI-generated lyrics in his style, he commented, "Songs arise out of suffering, by which I mean they are predicated upon the complex, internal human struggle of creation and, well, as far as I know, algorithms don't feel."
These diverse viewpoints highlight the ongoing debate within the music community. While some artists embrace AI as a new creative frontier, others worry about the potential loss of human emotion and experience in music creation. As AI technology continues to evolve, it's crucial to consider these perspectives and find ways to integrate AI that enhance rather than diminish the role of human artists.
Most AI systems train on existing works found on the Internet or through a library of music fed to it by humans. Training an AI model on a dataset of copyrighted music works without permission to do so is unethical, and we oppose it.
Most recently, two AI music companies, Suno and Udio, were sued by the RIAA for training their products on copyrighted music. This approach violates copyright laws and infringes upon the rights of artists and copyright holders. It devalues human creativity, creates a future where people undermine the original effort in music creation, and leads to the dehumanization of music.
To build genuinely ethical AI in music, we must prioritize collaboration between AI developers and copyright owners. We must partner with rights holders and artists to create high-quality datasets of music for training. In addition, track licensing should be a priority, and creators should be fairly compensated for their contributions.
Who Owns Copyright To AI Music?
As AI music grows in popularity and adoption, the copyright conundrum remains.
Who owns an AI-generated music track—the AI tool’s creator, the AI tool, or the artist who prompted it?
The existing structures of copyright law do not adequately address these new forms of generative AI. And there’s also little case for setting a precedent, which begs the urgent need for new legal frameworks.
The US Copyright Office’s current stance on AI-generated works is that human authorship is a bedrock requirement for copyright protection.
A policy statement issued by the Copyright Office in March 2023 stated, “When an AI technology determines the expressive elements of its output, the generated material is not the product of human authorship. As a result, that material is not protected by copyright and must be disclaimed in a registration application.”
This existing guidance implies that works created by AI without human intervention or involvement can not be copyrighted.
The Copyright Office also said that works containing AI-generated material may receive copyright protection if there’s sufficient human authorship. For example, if a human creates a unique prompt that creatively produces AI material, the material can still be protected by copyright.
In 2023, the USCO denied a copyright application for an award-winning image made by the generative AI image tool Midjourney. They cited a lack of human authorship necessary to support a copyright claim as reason. According to their logic, copyright protection should only apply to original works reflecting human creativity or input.
However, this position may shift as generative AI technology evolves and new legislation specific to AI-generated works emerges.
SOUNDRAW’S Approach to Ethical AI in Music
At SOUNDRAW, we are deeply committed to ethical music production in the evolving landscape of AI-generated content. Our approach addresses many of the concerns surrounding AI in music while embracing its creative potential.
In-House Production: Unlike platforms that train on copyrighted material without permission, we use music entirely produced in-house. This ensures that every track you generate is free from copyright infringement. Our AI system trains on sounds crafted by an in-house team of talented music producers, who meticulously create each sound to reflect quality and originality.
Balancing AI and Human Creativity: We believe in augmenting human creativity rather than replacing it. Our AI tools are designed to inspire and assist musicians and producers, not to render human input obsolete. This approach helps mitigate concerns about job displacement in the music industry.
Transparent Licensing: Thanks to our simple licensing model, users have full rights to use the music they create for personal or commercial projects. This transparency addresses legal concerns and provides peace of mind for creators using our platform.
Preserving Authenticity: By using our own sound library, we maintain a unique music identity. This approach helps combat the potential homogenization of music that could result from multiple AI systems training on the same datasets.
Ongoing Ethical Considerations: We continuously evaluate and adapt our practices as the AI landscape evolves. This includes staying informed about legal developments, engaging with artists and industry professionals, and considering the long-term implications of AI in music.
Global Perspective: While our approach is rooted in current copyright laws, we're mindful of how AI in music is perceived and regulated globally. We strive to create a product that respects international perspectives on music creation and ownership.
Looking to the Future: As AI technology advances, we're committed to finding new ways to ethically incorporate these developments into our platform. This includes exploring potential collaborations with artists and rights holders to expand our sound library while ensuring fair compensation.
Education and Transparency: We believe in educating our users about the ethical implications of AI in music. Our platform includes resources to help users understand how our AI works and the importance of ethical considerations in music production.
By taking this comprehensive approach, SOUNDRAW aims to lead by example in the ethical use of AI in music. We're excited about the creative possibilities AI offers while remaining committed to supporting human artistry, respecting copyright, and contributing positively to the music industry's future.
As the landscape of AI in music continues to evolve, we invite ongoing dialogue with artists, producers, and listeners. Together, we can shape a future where technology enhances creativity while upholding the values that make music a vital part of human culture.
Want to join us in our quest to maintain ethical AI in music?
