Product
The engine that turns project facts into filing-ready copyright language.
RightsDocket handles the filing logic behind AI-assisted music so you can see what to claim, what to exclude, and what to fix before you file.
50+ filing decisions handled behind the scenes
3 eCO-ready outputs generated from the same underlying logic
Deterministic rule mapping, not chatbot-style legal copy
Built for AI-assisted music, not generic content forms
What you get
Not a taxonomy. A cleaner filing record.
You do not need to understand the whole decision tree. You need a structured output that tells you what belongs in your claim, what must be excluded, and where your record is still weak.
Author Created
Structured language describing the human-authored material you can claim.
Material Excluded
Clear exclusion language for AI-generated material, third-party material, or anything that should not be claimed.
Note to Copyright Office
A concise explanation of the creative process, built from the actual facts of the project rather than generic AI prose.
How it works
Four steps from messy process to claim-ready language.
The engine exists to remove decision burden, not to create more of it.
Start with project facts
Contributors, AI tools, evidence, and work context go in. Nobody has to translate those facts into copyright jargon by hand.
Resolve claim logic
RightsDocket evaluates the scenario, separates human-authored material from AI-generated material, and determines what belongs in the filing record.
Generate filing-ready language
The engine produces the actual claim outputs the U.S. Copyright Office filing requires, plus readiness and risk signals.
Package a reviewable record
You get a structured, signed record that can be reviewed internally, shared with counsel, or used to support a filing workflow.
Why trust it
Built to be reviewable, not mystical.
RightsDocket is not a chatbot improvising legal-sounding copy. It uses structured claim logic to produce consistent, reviewable outputs from the facts the user provides.
What that means for you
- Reduces filing ambiguity before submission
- Standardizes claim language across projects
- Surfaces evidence gaps and review risk early
- Creates a cleaner internal record for labels, counsel, and collaborators
What it does not do
- It does not file with the Copyright Office for you.
- It does not turn weak facts into strong authorship.
- It does not replace legal judgment in edge cases.
- It does make the record cleaner, more consistent, and easier to review before you file.
Plain-English summary
RightsDocket's claim-preparation engine handles the filing logic behind AI-assisted music so you can see what to claim, what to exclude, and what to fix before you file.
Under the hood
How the claim engine works.
A transparent view of the filing logic behind every RightsDocket output. Most customers never need to see this - the engine handles it. But if you want to understand the rigor, or if you're reviewing RightsDocket output on behalf of a client, here it is.
Creation scenario classification
Every AI-assisted music project matches one of seven documented creation scenarios - from fully human compositions that used AI as a production tool, to works where AI generated core melodic and harmonic material that the creator then arranged and revised. The engine identifies which scenario applies and routes the claim logic accordingly.
7 scenario templates - each with distinct claim posture - deterministic routing from project facts
Musical element separation
The engine separates a work into its constituent elements - lyrics, melody, harmony, rhythm, arrangement, vocal performance, production - and classifies each as human-authored, AI-generated, or mixed. This separation determines what goes into Author Created versus Material Excluded on the eCO form.
8 element categories - human / AI / mixed classification - maps directly to eCO fields
Filing path and risk assessment
Based on the scenario and element classification, the engine determines the appropriate registration type (PA, SR, or both), flags risk conditions (incomplete disclosure, unseparable contributions, weak revision claims), and generates a readiness assessment with specific items to resolve before filing.
3 registration types - 7 risk categories - 3 legal precedents (Zarya, Rose Enigma, Feist) coded into the engine
Deterministic template assembly
The claim language is assembled from rule-mapped templates, not generated by an LLM. The same inputs always produce the same outputs. Every phrase in the Author Created, Material Excluded, New Material Included, and Note to Copyright Office fields traces back to a specific user input and a specific rule - reviewable, auditable, consistent across projects.
Zero LLM in the claim pipeline - same inputs -> same outputs - every phrase traceable to a rule
Why this transparency matters: If a USCO examiner requests additional information about how the claim language was prepared, the creator can point to a deterministic, rules-based system - not a chatbot that improvised something plausible. That distinction matters in legal review.
See the engine in context.
Start with the sample report if you want to see the output. Start with the analyzer if you want to test the first step yourself.