Music groups published seven AI principles together

More than 40 organisations signed the Human Artistry Campaign's principles requiring consent and licensing for AI training.

Historical event

March 16, 2023

First source published

March 16, 2023

Site publication

September 18, 2026

What happened

On 16 March 2023, at a SXSW event in Austin, a coalition of more than 40 organisations launched the Human Artistry Campaign. The RIAA's own announcement, from one of the founding members, describes the coalition as including major unions, trade associations and policy experts representing performers, writers, athletes and other creators, and names the launch event's speakers, including songwriter Dan Navarro and singer-songwriter Jessy Wilson.

What the documents say

The campaign published seven Core Principles for Artificial Intelligence Applications, current as retrieved on the campaign's own site on 16 September 2026 and matching the text of the 2023 launch announcement. The principles state that "use of copyrighted works, and the use of voices and likenesses of professional performers, requires authorization and free-market licensing from all rightsholders," that governments "should not create new copyright or other IP exemptions that allow AI developers to exploit creators," and that "copyright should only protect the unique value of human intellectual creativity." A coalition statement of this kind is evidence of what its signatories publicly committed to advocate, not evidence of a change in law; none of the principles has the force of a statute or a court ruling on its own. Later developments, such as Sony Music's 2024 opt-out declaration and the EU AI Act's 2025 disclosure obligations, run in the same direction as the principles but were separately negotiated and enacted, not automatic results of the campaign.

Why it matters for makers

The principles put a specific position on record: consent and licensing, not a blanket exemption, should govern whether a system trains on a creator's work. For an independent musician deciding whether to sign a platform's terms or push back on them, the campaign's language, particularly "free-market licensing from all rightsholders," is a template for what a licensing ask can look like, even though the campaign itself has no enforcement power.

What to check before you use it

Signing on to a coalition's principles is advocacy, not a licence or a contract; check what a specific platform's own terms actually say about training data and consent rather than assuming the industry position has changed the platform's practice. This is an editorial reading: a principle stated publicly by a trade body is a starting point for negotiation, not a right that arises automatically.

• Does a platform I am considering describe its training data policy consistent with these principles, or does it rely on a fair-use argument the principles reject?
• Which organisations relevant to my own work, if any, are signatories, and have they taken further action beyond signing?
• Has a law or ruling actually implemented a principle I am relying on, or is it still only advocacy?

A seven-point statement from 40-plus organisations shows where an influential part of the industry stood in March 2023; it is a useful benchmark against which to read every subsequent settlement, statute and ruling, not a substitute for any of them.

Sources & reading trail

Founding member's own announcement of the coalition's launch, size, principles text, and SXSW event.

Source published: 16 March 2023 · Retrieved: 16 September 2026

The campaign's own current statement of its seven core principles, matching the 2023 launch text.

Source published: Not established · Retrieved: 16 September 2026

Papers, terms and official documents establish the record; the maker reading and the checks are Signal to Song editorial analysis. This retrospective draft does not imply the site published on the event date.

Continue reading

• Sony Music declared an opt-out from AI training everywhere
• Record labels sued Suno and Udio over training data
• A nonprofit began certifying AI models trained on licensed data
• Browse the complete the archive

Published September 18, 2026, not on the date of the event described.