by Ana-Maria BACIU

Undoubtedly, the digital landscape of music is currently undergoing a profound transformation, arguably the most impactful since the invention of the phonograph. Today, however, the needle isn’t hitting vinyl; it’s hitting an algorithm. The intersection of Artificial Intelligence (AI) and Intellectual Property (IP) has moved from theoretical debate in law journals to high-stakes litigation in federal or national courtrooms.

At the heart of this storm stands a fundamental question:

When a machine composes a melody, who—if anyone—owns the song?

The Cultural Ignition: Authenticity vs. Algorithms

I have recently had several opportunities to speak, write and debate about one of the most visible “AI artists” in Romania – Lolita Cercel.

With songs that reached almost four million views on Youtube alone, and being described as “a human-curated digital music project, where AI is the tool, not the story”, Lolita has sparked vivid debates not only about AI generated music but also about cultural appropriation in a world that seems to be more about consumption and less about traditions.

In response to increasing visibility of AI generated music, one of the highest ranked Romanian artists, Delia, has recently launched a song explicitly named Analog that is intended to bring to the fore the authenticity of emotions and the need for real connections in a world that is constantly changing, dominated by digital and the expansion of artificial intelligence.

From Prompts to Playlists: The AI Music Revolution and Its Existential Crisis

The rise of generative AI platforms like Udio and Suno has democratized music production, allowing users to generate high-fidelity tracks from simple text prompts. These tools can replicate specific genres, moods, and even the vocal "vibe" of established artists. While this fosters a new era of creativity, it has triggered an existential crisis for the music industry.

The legal tension exists in two distinct phases:

• Input, the data used to train the models. Major labels argue that training AI on copyrighted catalogs without permission is "mass infringement," while tech companies often cite "fair use," arguing that the AI is learning patterns rather than copying specific works.
• Output, the music the models generate, raising questions about whether this “generated music” infringes upon or dilutes the value of the original training data.

A Tale of Two Strategies: Litigation vs. Licensing

The music industry is currently split on how to handle this "humanity" requirement.

• The Aggressive Path: Sony and Universal Music Group have pursued aggressive litigation against AI firms, seeking billions in damages for unauthorized training. They view AI as a potential "dilutive substitute" for human artists.
• The Collaborative Path: In contrast, Warner Music Group (WMG) has pivoted toward a "licensed future." Thus, according to publicly available information, WMG will begin an artificial intelligence music venture with technology start-up Suno - a year after it sued the firm in a landmark case. As part of the settlement agreement, WMG will let users create AI-generated music on Suno using the voices, names and likeness of artists who opt-in to the programme. I would argue that this approach represents a paradigm shift towards monetizing AI-generated content through licensing rather than attempting to fight its inevitable expansion.

Global Divergence: The EU AI Act

While the U.S. relies heavily on court precedents, the European Union has taken a regulatory approach, and the collision between AI and IP is being shaped not just by courtroom battles, but by a distinct legal philosophy that prioritizes the "moral rights" of the creator and the strategic regulation of the Digital Single Market.

While the US often views copyright through an economic lens, the European approach—steeped in the droit d’auteur tradition—centers on the personal link between the author and their work. This distinction has made the EU the primary battleground for the future of AI music.

Conclusion: The Future of the "Author"

We are entering a "hybrid era." The law is clear that a machine cannot be an author, but the line between a "tool" (like a piano or a digital audio workstation) and a "creator" (the AI) is blurring.

For musicians, the strategy is shifting from fighting technology to documenting their creative process. To secure a copyright in 2026, an artist must prove they were the "mastermind" behind the algorithm, not just a passive recipient of its output.

As the courts seem to agree, the law currently protects the human spirit. If you want to own the music of tomorrow, you must ensure your fingerprints are all over the code.