Claude General

Output Rights Prompts: 6 Before You Sell Something an AI Made

Suno just launched label-backed models, publishers are suing over training data, and a court has already ruled on how that data was acquired. If you sell work an AI helped produce, the question is what you can establish about it. These six prompts are for finding that out before a client asks. None of this is legal advice and none of it replaces a lawyer.

⌨️ 6 prompts 🕐 Updated Sep 11, 2026

💡 How to use these prompts: Replace everything in [BRACKETS] with your specific details before sending. Click Copy to copy any prompt to your clipboard instantly.

Read what the terms actually grant

Most disputes turn on a clause nobody read. Quoting rather than summarising is what makes this usable.

Here are the terms for a tool I use to produce client work: [PASTE]

Quoting the clause each time, tell me:
- Who owns the output
- Whether commercial use is permitted and at what scale
- Whether rights survive if I cancel
- Whether the provider indemnifies me against third-party claims
- What is not addressed at all

Say unaddressed rather than inferring. Silence is not permission.

Check what my client contract requires

The last line matters. Some contracts are incompatible with generated material and it is better to know first.

Here is the relevant part of a client contract: [PASTE]

Tell me:
- What I am warranting about ownership and originality
- Whether AI-assisted work satisfies that as written
- What would breach it
- What wording I would need changed to be safe

If the honest answer is that I should not use AI on this deliverable, say so.

Establish what I could prove

Written before a challenge this is diligence. Written after, it is a scramble.

I produced this using AI: [DESCRIBE THE TOOL, MODEL AND PROCESS]

If someone challenged its provenance tomorrow, what could I establish and what could I not?

Be blunt about the gaps. Tell me what record I should have kept and whether it is too late to reconstruct it.

Design a record I will actually keep

Short records survive. Comprehensive ones get abandoned in month two.

I produce client work with AI assistance regularly.

Design a provenance record capturing enough that in two years I could establish how any given deliverable was made, without relying on memory.

Keep it to fields I will realistically fill in every time. An elaborate schema I abandon is worse than a short one I maintain.

Ask the vendor the right questions

Indemnity decides who carries the risk. Most people never raise it.

I am choosing a tool for commercial production: [NAME OR DESCRIBE]

Draft the questions to ask about output ownership, training data, indemnification and what happens if terms change.

For each, tell me what a reassuring answer sounds like and what an evasive one sounds like. Mark which answers I should get in writing.

Work out whether to disclose

The last line prevents a reflexively cautious answer. Norms differ enormously by field.

I used AI to help produce: [DESCRIBE] for a client who has not asked about it.

Tell me:
- Whether disclosure is required by the contract
- Whether it is required by professional norms in this field
- What the risk is of not disclosing
- How I would phrase it if I chose to

Do not default to advising disclosure. Tell me what actually applies here.