TrackOrigin Launches World-First Live Test That Proves a Human — Not AI — Made the Music
Nyenta.com/10347572

NEW CITY, N.Y. - Nyenta -- TrackOrigin (trackorigin.io) today announced the beta release of a new standard for proving human-made music — one that abandons audio-only AI detection in favour of testing the living relationship between an artist and their work.

The launch arrives as the streaming catalogue tips into crisis. Deezer reported in April 2026 that it now receives nearly 75,000 fully AI-generated tracks daily — around 44% of uploads. A Deezer/Ipsos study of 9,000 listeners across eight countries found 97% could not reliably distinguish AI-generated music from human-made. When the audience can't hear the difference, the file alone proves nothing.

More on Nyenta.com

• Rebuilding investor confidence in quantum through better board education
• Frankie Sly Takes on New Role as Face of Global Creators' Rights Campaign
• UN Blockchain Week 2026: Priority Speaker Round Closes in 5 Days
• Registration Open for Long Island Crisis Center's 18th Annual Suicide Prevention Walk
• Fatal FOMO May be Your Last Roll of the Dice

TrackOrigin's answer is to stop guessing. Instead of asking a classifier whether a waveform "feels" human, it asks the artist to prove it — live.

An artist uploads a finished master (WAV, FLAC or AIFF). TrackOrigin fingerprints the audio, then runs a 60–120-second recorded session in which the artist demonstrates authorship: performing fragments on camera, recalling creative decisions only the author would know, and showing their process. Challenges are generated after upload from the track itself, so there's no answer key to rehearse.

The breakthrough is underneath. Multiple independent engines verify in parallel across acoustic, behavioural, visual, linguistic, cryptographic and adversarial evidence. A verdict requires convergence — several unrelated signals agreeing. To fake a certificate, an impostor would have to defeat all of them at once, in real time, for prompts they can't see in advance. That costs more than simply making the track.

More on Nyenta.com

• Raja Limousine Expands Premium JFK Airport to Manhattan Car Service in 2026
• WindowFixGuru™ Launches Special Savings on Wood Window Restoration in Nassau County
• Martin A. Sumichrast Joins Hawkeye Systems, Inc. as Chairman of the Board
• Lorecentric Inc. Announces Official Launch as a New York Corporation
• Bill Gray's Sponsors Kidtopia Summer Camp for Rochester-Area Children

A passing session produces a cryptographically signed certificate: an Ed25519-signed manifest bound to the audio's SHA-256 hash, a permanent public certificate page, and an embeddable Origin Seal. Change one sample and the hash changes, voiding the certificate. The proof travels with the work and verifies offline — without contacting TrackOrigin.

"For a century, hearing a song was enough to trust it. That's over," said a TrackOrigin spokesperson. "Listeners can't be the detector any more — 97% get it wrong. So we moved the proof off the file and onto the act of making it."

Crucially, AI use isn't disqualifying — it's declared. TrackOrigin records any AI assistance on the certificate as disclosed context; only fully AI-generated tracks are ineligible. The approach maps to emerging disclosure law, including EU AI Act Article 50 (enforceable 2 August 2026), China's GB 45438-2025, and California's SB 942.

👉 Try TrackOrigin Now (https://trackorigin.io/)

Media contact: support@trackorigin.io