Thecategory.

Every seat at the table

A music deal has more than two sides, and every seat needs something different from the same agreement.

Artists and songwriters

Understand what you are signing, preserve what you own, and return to the terms when money or rights become unclear.

Managers and agents

Run agreements, documents, deadlines, and evidence across an entire roster without rebuilding every deal from scratch.

Labels and publishers

Coordinate contracts, catalog information, negotiations, approvals, and ongoing obligations across a team.

Business affairs

Clear deals across a whole catalog on consistent positions, with every negotiation on the record and the signed terms ready when finance or a royalty query asks for them.

Venues and promoters

Prepare and execute booking or performance agreements while keeping dates, requirements, and final documents connected.

Counsel and legal teams

Start from the source language and a graded issue trail, keep every negotiation decision on the record, and close with a signed, sealed agreement. The judgment stays yours; the first read and the paperwork do not have to.

A tool built for one of those seats hands the deal to the next in its own format, and the reasoning stays behind. WALDHORN.AI is built for the table rather than a chair: each seat works from the same agreement, the same issue trail and the same signed copy, and sees what its role needs.

The hard part is what happens next

Most tools stop at “here is what your contract says.” That is the easy half. The value is in what happens next, and almost no one carries it through.

WALDHORN.AI reviews the agreement clause by clause, proposes the changes, runs the negotiation, takes the signature and seals the signed copy, drafts the next agreement, carries the rights, splits and deadlines forward, and checks every royalty statement against the terms that were signed. A reader that only reads leaves you where you started: informed, but not equipped.

The signature is where a music deal starts working, not where it ends. The statements arrive for as long as the recording earns, and the law it is written under keeps deciding things for decades: when a term can no longer be enforced, how long an artist has to audit, when a grant can be taken back. Carrying the deal the whole way is harder to build and far harder to copy, because every step has to speak the language of the one before it.

Where everyone else sits

The legal-AI field is real and growing, but it is aimed elsewhere. General chatbots answer anything and keep nothing. The assistants built for lawyers make attorneys faster at reading and drafting. The enterprise contract platforms take a corporate contract from draft to signature, and do it well. Music has its own specialists, each holding one stage: contract analysis, rights and metadata, royalty accounting.

The gap is not an accident of who built first. It is structural. The general platforms follow a corporate contract to its signature, and some keep its deadlines and files after it. A music deal needs far more there: the splits, the clearances and the statements, for as long as the recording earns, read against terms that differ from one city to the next. Serving that takes a depth in music that a general tool has no reason to build, and a reach across stages that a one-stage specialist was never built for.

Depth a general tool has no reason to build

Reach is half of it. The other half is depth. A general tool reads a music contract as it reads any other, and a music contract is not any other: it has its own agreements, its own traps and its own law, place by place. This is what WALDHORN.AI brings to each deal before a word is drafted or a clause is graded.

It keeps no stock of contracts, and that is the point. A library of ten thousand agreements is ten thousand other people’s deals, signed under other laws; it cannot say what the law requires where you sign, or what your deal needs. WALDHORN.AI asks about the deal in front of it and drafts that one.

38

kinds of music agreement it drafts

2,577

questions it can put about a deal, up to 118 for a single one

60

places whose contracts get verified rules, and every rule dated

456

known traps in the review's playbook, written by hand

Behind those figures is a ledger of the law in the 15 jurisdictions read so far, every rule tied to the statute or judgment behind it and dated when a person last checked it. The engine essay sets out what is done with it; the register lets anyone check it.

A market that has already arrived

Two things arrived at once. After a decade of very public disputes, the people who sign music deals know a contract can make or break a career. And language models now write so fluently that a wrong answer reads exactly like a right one, which makes an answer that can be checked worth more than it has ever been.

Global recorded-music revenue reached $31.7 billion in 2025.1 U.S. recorded-music wholesale revenue reached $11.5 billion in 2025.2 Who is paid which share of it is settled by agreements people signed, and the law that governs each one depends on where it was signed. WALDHORN.AI is not chasing the enterprise centre of legal AI. It starts where the general tools do not reach: music's own agreements, and everything they go on deciding after the signature.

Why music first

Music is the entry point because it is where the asymmetry is oldest and the deals follow a career the longest. A songwriter signs at twenty-three, and at fifty-eight the law hands them the choice of taking the grant back (Figure 1). That makes it the hardest version of the problem rather than the easiest, which is exactly why it is the right place to prove the work.

It is not the ceiling. The same asymmetry runs through every field where people sign away rights they do not fully understand, and depth in one field is what makes the next one reachable at all.3

The invitation

For artists, managers, labels, publishers and their counsel, WALDHORN.AI is a safeguard to use today, at the moment it matters: before the signature, not after the dispute.

The bet is simple. In a few years every music deal should arrive already read against the law of the place it is signed, and stay readable for as long as it pays. We intend to be the workspace where that happens.