AI Audio Protection Standard
OverviewWhat is AIAPS?
AIAPS, the AI Audio Protection Standard, is a public specification that lets musicians and rights holders declare that a recording is not authorized for AI training. It combines proven techniques from the music industry into a single verifiable declaration that travels with the audio. Every protected track carries a timestamped record like this one, establishing the rights holder's stated terms for any legal framework, platform, or licensing system to reference.
AIAPS Protection Certificate
VerifiedAIAPS Protected — NO_AI_TRAINING
Usage Policy
This recording is registered under the AI Audio Protection Standard (AIAPS) and is designated as not authorized for AI training, dataset ingestion, voice cloning, style modeling, or synthetic generation without permission from the rights holder.
AIAPS Registry — Verified Record
Rev. 1.0
Section 01 — Precedent
AIAPS Creates a Record. It Does Not Block AI.
It follows a long line of public signaling systems. They don't enforce. They establish the terms. Everything else references them.
01
robots.txt
Does not block crawlers. Establishes a published preference that well-behaved systems honor and bad actors are on record for ignoring.
02
Copyright notice / ISRC
Does not prevent copying. Establishes authorship and a referenceable record that legal and industry systems can cite.
03
"No Trespassing" sign
Does not physically stop anyone. Changes the legal and social meaning of the act.
04
AIAPS
Creates a public, timestamped, cryptographically verifiable record of your stated terms. Any future framework (legal, regulatory, platform policy, licensing) can reference it.
The more recordings on record, the stronger every declaration becomes. Getting on record now means your terms predate the frameworks that will reference them.
Section 02 — Four Components
Four Components
AIAPS brings together techniques that are already widely used throughout the music industry. Any of these four can be used to verify a protected recording.
Component 01
Audio Fingerprint
A perceptual fingerprint derived from the spectral structure of the recording. Designed to survive lossy compression (MP3, AAC, OGG) and used to identify and match the audio in the AIAPS registry.
Component 02
Embedded Watermark
An inaudible watermark embedded in the audio samples, encoding the track's AIAPS ID and NO_AI_TRAINING policy. Preserved in lossless distribution formats (WAV, FLAC, AIFF).
Component 03
Registry Record
Each AIAPS-protected recording receives a unique AIAPS ID and a timestamped registry entry linking the fingerprint, watermark payload, and rights-holder declaration.
Component 04
Metadata Notice
An AIAPS-PROTECTED notice written to the file's metadata tags where supported by the format, providing a human-readable and machine-readable declaration that the recording is not authorized for AI training.
Adopt the Standard
Download the AIAPS Plugin
The AIAPS Plugin is the official implementation of AIAPS v1.0. Protect, register, and verify your recordings under the standard. Now in early access for macOS.
macOS • AU, VST3, Standalone