How does a non-musician end up building a copyright tool?
It started with a text from my girlfriend. She sent me a YouTube video — a song about recovery. It was visceral, beautifully produced, and emotionally precise in a way that felt almost impossible. She didn't know it was AI-generated. She just knew it moved her.
I felt something else entirely. After twenty years in regulated industries — pharma, MedTech, orthopedics — I've been trained to look for one thing above all else: the audit trail. In those worlds, "truth in labeling" isn't a suggestion. It's enforced reality. So I started pulling on the thread, and that thread led me to Suno.
Within minutes, I experienced two things simultaneously: the miracle of creation and the complete absence of proof.
Why would someone use AI to make music about recovery?
I'm not a musician. I'm a 1.5-generation immigrant who went from Kolkata to Boston. In the South Asian community, addiction isn't discussed openly. It's framed as a moral failure, a weakness, a sin. For over a decade in recovery, I tried to communicate what I'd been through — what it actually is, biologically and emotionally — to the people I love. Speaking didn't work. Writing didn't work. The words were there, but the medium couldn't carry the weight.
AI music changed that. I created an album called Twelve Steps from Kolkata to Boston, and one song in particular — "The Glitch (Not Your Fault)" — was my attempt to explain something specific to my family: addiction isn't a failure of character. It's biology. It's a glitch in the wiring, and it's not your fault for having it.
I wrote every word of those lyrics. I shaped the story, the emotional arc, the sequencing. AI helped with production, instrumentation, and voice rendering. The result wasn't purely human and it wasn't purely machine. It was something in between — something that carried a truth I'd spent years trying to articulate.
That's the real power of AI music tools. They don't replace human expression. They break down barriers to it.
People who couldn't afford studios can now produce. People who couldn't articulate emotions can now communicate them. People who felt invisible can finally create something that feels seen. But that power means nothing if creators can't protect what they made.
What happens when you export an AI-assisted track?
When I exported my tracks, every file looked identical to something generated by a bot with zero human input. No metadata explaining what was human versus AI. No chain of custody. No structured disclosure. No proof I could defend. Just a bare audio file that carried all the legal weight of a screenshot.
And that's when the real question hit me: how do I prove this is mine?
The U.S. Copyright Office is clear — only human-authored elements are protectable. Lyrics, arrangements, creative decisions about structure and sequencing: those are yours. Pure AI output is not. But here's the problem: there is no system that maps your creative process into something the Copyright Office actually understands.
Most creators are guessing at those fields. Many don't even know they exist. So even when you have legitimate human authorship — even when you wrote every word, shaped every arrangement, made every creative decision that matters — you can't prove it cleanly. The documentation gap between creation and registration is where rights go to die.
What does RightsDocket actually do?
I built RightsDocket because I needed it. Not as a business idea. As a missing layer.
RightsDocket is the first and only tool purpose-built for documenting human authorship in AI-assisted music and generating USCO-ready claim language. It walks creators through a guided workflow: upload your track, document your contributors, disclose which AI tools you used and how, review your copyright readiness score, and generate the exact Limitation of Claim language — Author Created, Material Excluded, and Note to CO — that the Copyright Office requires.
This isn't a template you fill in and hope for the best. We mapped the actual USCO Limitation of Claim taxonomy, ISRC standards, and key legal precedents — Zarya of the Dawn, Rose Enigma, Feist Publications — into 56+ decision nodes that determine what goes into each field. The output is a Provenance Pack: a signed PDF and JSON bundle containing your wizard data, USCO form mapping, risk flags, contributor table, creative decision log, and filing pathway guidance.
The tool that helps you prove human authorship can't itself rely on a black box.
The claim language is deterministic — rule-mapped and template-assembled from your documented inputs, not generated by AI. That's a deliberate design choice that reflects the core trust constraint of the entire platform.
You made something with AI. Now prove it's yours.
AI didn't remove human creativity. It amplified it. It let me say something to my family that a decade of recovery couldn't unlock. But the systems that protect that creativity haven't caught up to the tools that enable it.
That's what RightsDocket is for. Not just for me. For every creator who used AI to make something real — and needs the documentation to prove it.