Jen AI
Filed · ethically trained
Open Jen
Filed § 2026-06-09 Ethically-trained AI music ¹

Music with receipts.
And stems.

Jen AI is the generative music platform from Futureverse, trained only on 40+ fully-licensed catalogs. Type a prompt, get a multi-stem track. Continuation, in-painting, R3IMAGINE™. Every output gets a JENUINE™ certificate on the Root Network.

Licensed catalogs only 4-stem export On-chain provenance Instrumentals¹
Procedure § I–III From prompt to filed track ²

Three stages.
One filed track.²

Jen's pipeline is built around one idea: every track is a document, and documents have provenance.

STAGE · IPROMPT
01·a

Write the brief.

Type a prompt — "warm jazz quartet, brushed snare, slow swing, B♭ minor" — or upload an existing track to R3IMAGINE™. Optional: pin genre, mood, BPM, instrumentation. The shorter and more specific, the better the first take.

STAGE · IIRENDER
02·b

Get four stems.

The waveform-diffusion model renders a stereo track in seconds. You receive not just a mixdown but four separated stems — drums, bass, melody, additional instrument — ready to drop straight into your DAW.

STAGE · IIIREFINE
03·c

In-paint and continue.

Don't like bar 9? Use in-painting — re-prompt just that section. Need 90 more seconds? Use continuation — Jen extends the track in style. Every regeneration filed on the Root Network with a fresh JENUINE™ certificate.

§ IV — Signature The JENUINE™ stamp ³

Music with receipts.
Every time.

Most generative music platforms train on scraped catalogs and dare you to figure out the copyright. Jen does the opposite. Every track Jen generates is checked against 150 million reference tracks for similarity, then issued a cryptographic hash and an XLS-20 NFT on the Root Network — a public record that this track was Jen-generated, when, by which account, from what prompt.

  • Licensed-only training — Jen's model trained on 40+ fully-licensed catalogs. No scraped data, no "fair use" claims, no take-down risk.
  • 150M-track similarity check — every output is screened against the same database major labels use. If something accidentally matches a copyrighted piece, you'll know before you ship.
  • JENUINE™ certificate — verified outputs receive a JENUINE™ indicator, signed cryptographically on the Root Network with hash + timestamp + your FuturePass.
  • XLS-20 NFT for IP rights — each track gets an on-chain token that follows you when you sell, license, or transfer the work. The provenance is portable.
Run the filing
CERTIFICATE OF PROVENANCEVERIFIED ON-CHAIN
JENUINE VERIFIED

"Warm jazz quartet · brushed snare"

Composition Schematic · No. JEN-2026-417

Hash0x4f7a8c…2e3b91
TokenXLS-20 #4172
Filed2026-06-09 · 14:22:08 UTC
FuturePass0x7a2c…f481
Similarity check0/150M · clear
Length2:48 stereo · 4 stems
ROOT NETWORK · BLOCK 8,294,617 ↗ ON-CHAIN
Applicants Who files with Jen

Four kinds of
music applicant.

Jen's audience splits by how much they care about the provenance trail. The more you ship, the more it matters.

No. 01
SY

Sync & licensing

Film, TV, ad agencies. People whose lawyers ask for clearance paperwork. JENUINE certificate + on-chain hash means clean sync without IP detective work.

No. 02
PR

Producers shipping

Working producers using AI for backing tracks, b-sides, demos. They need stems, not just mixdowns. They need provenance, not Suno's vibes.

No. 03
CC

Content creators

YouTubers, podcasters, course creators who got hit by a Content ID claim once and never want it again. License paper trail per upload.

No. 04
LB

Labels exploring AI

A&R curious about generative tools but worried about the legal mess. Jen's industry-aligned approach is the path that doesn't end in court.

Schedule of features What's wired in

The full filing.

A short list of what Jen does once the alpha is open. Some features are flagged as ™ — they're Jen's proprietary research.

No. 1 · Multi-stem render

Four stems. One render.

Drums, bass, melody, additional instrument — each as its own audio file, time-locked. Drop them into your DAW and re-mix from scratch. Most generators give you a mixdown; Jen gives you the session.

DRUMS
BASS
MELODY
ADDIT.
No. 2 · Continuation

Extend the track.

Half a song? Upload it. Jen reads the existing audio and writes a plausible continuation in the same key, tempo, and feel.

No. 3 · In-painting

Replace a section.

Mark bars to redo, re-prompt. Jen re-renders that section, leaves everything else untouched. DAW-level surgery without leaving the browser.

No. 4 · R3IMAGINE™

Transform a take.

Upload an existing track and have Jen reinterpret it — same composition, different genre, instrumentation, or mood. Stems out, ready to mix.

No. 5 · StyleFilter™

Steer the model.

Bias the output toward a chosen style without giving up the prompt detail. StyleFilter is Jen's way to "nudge" without re-prompting from scratch.

No. 6 · On-chain provenance

Every track, filed.

Every Jen output gets a cryptographic hash + XLS-20 NFT on the Root Network. You don't have to do anything — the certificate ships with the audio. Sync agents and label legal teams can verify in one click.

JENUINE™Root NetworkXLS-20 NFTAuto-filedPortable
Comparative filing Honest comparison

Where Jen fits
— honestly.

Jen is the IP-clean, multi-stem, instrumental-only end of the AI music spectrum. It doesn't sing, and it isn't the cheapest. Here's the actual map.

Jen AI Suno Udio MusicLM Splice
Trained only on licensed data Yes — 40+ catalogs Disputed Disputed Google research Curated samples
Multi-stem export (drums/bass/melody/inst) Yes — signature Mixdown only Mixdown only N/A Stem packs
Vocals Instrumentals only Yes — its core Yes Vocaloid-style Sample packs
On-chain provenance / cert JENUINE + XLS-20 No No No Licence file
Continuation & in-painting Both Extend Extend N/A N/A
Free tier Alpha access Daily credits Daily credits Research demo Subscription
Honest: if you need vocals, Jen is not your tool — Suno and Udio do that, Jen doesn't. If you want the cheapest, casual song generator, again not Jen. Jen wins when (a) you need stems not mixdowns, (b) you need a verifiable licence-clean paper trail, and (c) you're shipping the music commercially. The product is paid in alpha; pricing details on jenmusic.ai.
Affidavits What applicants report

Three takes,
one mixed review.

From producer Discords, sync forums, and content-creator Slacks. The 4-star one stays in.

★ ★ ★ ★ ★ No. 01

"Sent the JENUINE certificate to our label legal. They cleared the sync the same day. That's the first time AI-generated music has gone through approval without a meeting."

M
Maya O.Sync supervisor · LA
★ ★ ★ ★ ☆ No. 02

"No vocals is the dealbreaker. I came in expecting Suno-with-receipts. Stems are great. Quality is great. But I need a singer."

D
Devon T.Producer · Atlanta
★ ★ ★ ★ ★ No. 03

"In-painting saved a podcast intro for me. Bars 5–8 were wrong, re-prompted just that section, rest of the bed stayed put. That's the right interaction model."

S
Sara V.Podcast editor · Berlin
Origin filing Provenance

A music platform
built like a patent office.

Jen is the generative-music product of Futureverse, an Auckland-based AI company co-founded by Shara Senderoff (named to Rolling Stone's "Future 25" and Billboard's "40 Under 40," previously co-founded blockchain venture fund Raised In Space with Scooter Braun) and Aaron McDonald. Founding partner: Mike Caren. The research foundation is JEN-1, a waveform-based latent diffusion model whose papers have been cited by Meta FAIR, Sony AI, and Bytedance AI.

The design choice that defines Jen is the refusal to scrape. While competitors trained on whatever was on the open web and worried about consequences later, Jen committed early to licensed catalogs only — over 40 partnerships before launch. The Root Network and FuturePass are how that commitment becomes machine-readable: every track gets a hash, a timestamp, a token, and a public record that no one disputes later.

What we'd flag honestly: Jen does not currently generate vocals — outputs are instrumentals only, by design. The product is in alpha, so feature pace and quality will keep changing. There's a learning curve around FuturePass, Root Network, and XLS-20 NFTs that's friction for users who just want music. And Jen is a paid platform; pricing details live on the official site.

For the official product, current pricing, and the full Futureverse ecosystem, see jenmusic.ai.

FILING LOGNo. JEN-2026-417
MakerFutureverse
Co-foundersSenderoff · McDonald
PartnerMike Caren
ResearchWaveform diffusion · JEN-1
CitationsMeta FAIR · Sony AI · Bytedance
Training40+ licensed catalogs
Output4 stems · stereo
VocalsNot yet
ProvenanceRoot Network · XLS-20
StageAlpha · paid
Interrogatories Frequently asked

Things people
actually ask.

What is Jen AI?
Jen AI is a generative-music platform from Futureverse that turns text prompts into multi-stem instrumental tracks. The model uses proprietary waveform-based latent diffusion, was trained on 40+ fully-licensed music catalogs, and produces stereo audio with separated drums, bass, melody, and additional-instrument stems. Every output is screened against 150 million reference tracks for copyright similarity and issued a JENUINE™ certificate with a cryptographic hash and XLS-20 NFT on the Root Network blockchain.
How is Jen different from Suno and Udio?
Three structural differences. (1) Training: Jen trained only on fully-licensed catalogs; Suno and Udio's training data is disputed and litigated. (2) Output format: Jen gives you four separated stems per generation; Suno and Udio give a mixdown. (3) Provenance: every Jen output ships with a JENUINE certificate and on-chain hash; Suno/Udio outputs do not. The trade-off: Suno and Udio generate vocals, Jen does not — it's instrumentals only.
Does Jen really not do vocals?
Correct, by design. The official site states: "Unlike its competitors, Jen currently only offers instrumentals — vocals are not available." This may change as the platform evolves, but as of the alpha, all Jen outputs are instrumental. If you need vocals, you'll need a separate tool (Suno, Udio) or a singer.
What's the JENUINE certificate, really?
It's a cryptographic record on the Root Network blockchain proving that a specific track was generated by Jen, at a specific time, by a specific FuturePass account, from a specific prompt. The certificate includes a SHA-style hash of the audio, a timestamp, and an XLS-20 NFT token that follows the track if you sell or license it. The purpose is to give sync supervisors and label legal teams a one-click verifiable provenance trail without manual paperwork.
Can I use Jen outputs commercially?
Yes. The training corpus is fully licensed, the similarity check against 150M tracks happens automatically, and the JENUINE certificate is your portable proof of clean provenance. Use Jen tracks in sync, advertising, streaming, content reels, podcasts, and games. Always check current terms on jenmusic.ai before a major commercial release.
What's continuation and in-painting?
Continuation: upload an incomplete track (yours or generated) and Jen extends it in the same key, tempo, and feel. Useful when you have a strong four-bar intro but no idea where the song goes next. In-painting: mark specific bars to redo and re-prompt just that section — Jen replaces only the marked region, leaving the rest untouched. Both are signature features missing from most competitors.
What's R3IMAGINE™?
R3IMAGINE is Jen's term for re-interpretation. You upload an existing track (yours or one with rights) and Jen reinterprets it in a different genre, mood, or instrumentation while preserving the underlying composition. Useful for sketching alternative arrangements of a tune you've already written, or for testing what your song would sound like as e.g. a lounge piece rather than a rock track.
Do I need a crypto wallet to use Jen?
Not in the conventional sense. Jen uses a FuturePass account, which abstracts the blockchain interaction — you sign up with normal credentials and the XLS-20 NFT minting happens behind the scenes. You don't manually approve transactions or pay gas. But the on-chain part is real: if you want to inspect your track's provenance entry on the Root Network explorer, you can.
What length of track can Jen generate?
Base generations are around 40 seconds of stereo audio with four stems. Using continuation, the platform can extend tracks to about 3 minutes — long enough for sync placements, content, and most production needs. The waveform-diffusion model targets quality over length; if you need a 7-minute prog-rock epic, you'll be doing several continuation passes.
Closing § X Open the platform

File a track.
Get the stems.

Multi-stem AI music with licensed-only training, on-chain provenance, continuation, in-painting, and JENUINE™ certificates. Instrumentals, in alpha, paid.

Futureverse · waveform diffusion · 4 stems · JENUINE™ · Root Network · alpha