Nestbitt Inc. ("Nestbitt", "we", "us") operates an AI-powered creative platform. This AI Disclosure & Synthetic Media Statement supplements the Terms of Service, the Acceptable Use Policy, and the Licensing & Usage Rights policy. It is intended to be transparent about how generative AI shapes what you create on Nestbitt and the risks involved.
How Nestbitt generates content
Music, vocals, lyrics, and cover artwork on Nestbitt are produced by artificial intelligence models in response to your prompts, settings, and inputs. These outputs are synthetic: they are generated by statistical models, not recorded by human performers or composed note-by-note by a person, unless you supply such material yourself.
- Music and vocals are generated by audio models that synthesize instrumentation and singing from text and parameter inputs.
- Lyrics and text may be generated or assisted by language models.
- Artwork (such as cover images) is generated by image models.
- Voice synthesis and cloning features reproduce or generate vocal characteristics and are subject to strict consent rules described below.
Synthetic-media labeling and transparency
We label AI-generated media where we provide it and support transparency about synthetic content. You must not remove, obscure, or falsify labels, watermarks, or metadata that identify content as AI-generated, and you must not present synthetic content as a genuine recording of a real person or event in a deceptive way.
The EU AI Act establishes transparency and labeling duties for providers and deployers of AI systems that generate synthetic audio, image, video, or text content under Article 50, with those obligations becoming applicable on 2 August 2026. Today we disclose AI-generated content on Nestbitt through three concrete mechanisms: (i) visible "AI-generated" indicators displayed alongside tracks and creator profiles in the application; (ii) structured-data markup that identifies the Nestbitt AI music-generation system as the creator of each recording in our public song pages, so search engines and other indexers can read the synthetic-content signal directly; and (iii) embedded provenance metadata on every audio download. The download metadata layer has two parts: text tags written into the file's container metadata (ID3v2 for MP3, MP4 atoms for M4A, Vorbis comments for FLAC and OGG, INFO chunks for WAV) carrying a per-track watermark identifier, a composer and publisher tag, and a link to this disclosure; and a signed C2PA Content Credentials manifest, which cryptographically binds the AI-generation declaration to a hash of the audio bytes so any subsequent tampering is detectable by independent verifiers. We will continue to refine these mechanisms — including adoption of any specific machine-readable provenance formats prescribed by future Commission guidance — as the final Article 50 implementing rules are published. This statement will be updated as the requirements and our implementation are finalized; nothing here should be read as a claim of certified compliance.
Conduct that misuses synthetic media — including deceptive deepfakes and civic manipulation — is prohibited under the Acceptable Use Policy.
Verifying a Nestbitt provenance manifest
Every audio file you download from Nestbitt carries a signed C2PA Content Credentials manifest declaring the file as AI-generated by the Nestbitt music-generation system. You can inspect it without trusting us: drop the file into Adobe's Content Credentials Inspector at contentcredentials.org/verify, or run `c2patool inspect <file>` locally using the open-source c2patool. The manifest binds our declaration to a cryptographic hash of the audio bytes, so any subsequent edit to the file invalidates the signature in a way verifiers can detect.
We sign manifests with a Nestbitt-controlled signing key. The corresponding public certificate is published at nestbitt.com/.well-known/c2pa-public-cert.pem, where any verifier can fetch it to confirm a manifest was issued by us. At present this is a self-signed certificate, so inspector tools may display a “untrusted signer” warning until an EU-recognised C2PA trust list emerges and we re-sign with a certificate chained to that list. The cryptographic signature itself remains valid in either state — the warning concerns the trust chain, not the integrity of the manifest. If we ever rotate the signing key, the previous public certificate will remain published indefinitely so older downloads remain verifiable, and any compromise will be disclosed on this page.
Copyrightability of AI outputs
Under current guidance from the U.S. Copyright Office (as reflected in its 2025 reports on copyright and artificial intelligence), material that is generated purely by artificial intelligence — without sufficient human authorship — is generally not protected by copyright. Human creative contribution, such as meaningful selection, arrangement, and modification, can affect whether and to what extent a resulting work is protectable.
Because purely AI-generated output may not be protectable, Nestbitt does not and cannot grant you exclusive ownership of, or exclusivity in, the raw AI outputs you generate. Similar or identical outputs may be generated for other users. The license you receive to use your generations, the commercial-use tiers, and the limits of those rights are set out in the Licensing & Usage Rights policy.
Training data and your responsibility
The AI models used on Nestbitt are trained on data assembled by us or by our model providers. We do not warrant the complete provenance, licensing status, or rights-clearance of every element that may have influenced a model, and the legal treatment of AI training data is unsettled and evolving.
You are responsible for how you use the outputs you generate. You bear the risk that a particular use may infringe a third party's rights, and you are responsible for clearing any rights and obtaining any consents necessary for your intended use — particularly for commercial release, public performance, or distribution.
We note that there is ongoing litigation in several jurisdictions concerning the use of copyrighted works to train AI music and other generative systems. These matters are unresolved, and we make no representation about how they will be decided or about how any outcome might affect AI-generated content. This is an evolving area of law.
Output similarity, accuracy, and hallucination
Review outputs before any commercial use
AI-generated outputs may unintentionally resemble existing works, names, voices, or styles, and may contain factual inaccuracies, fabricated content ("hallucinations"), or unexpected results. Outputs are provided AS IS, without warranty of originality, accuracy, non-infringement, or fitness for any purpose. You must review and, where appropriate, independently clear any output before relying on it, publishing it, or using it commercially.
Because outputs are probabilistic, identical or similar prompts can produce content that resembles other generations or pre-existing material. You are responsible for assessing this risk for your intended use.
Voice synthesis and anti-impersonation
Voice cloning and synthesis may only be used on voices and likenesses you are entitled to use. You must have the explicit, informed, and verifiable consent of any real person whose voice or likeness you reproduce, or another lawful basis (such as your own voice or a validly licensed voice).
Non-consensual voice cloning, impersonation, and deceptive use of synthetic voices are prohibited under the Terms of Service and the Acceptable Use Policy. Misuse may result in removal of content and termination of your account.
Human oversight and moderation
We combine automated safety systems with human review to detect and act on prohibited content and misuse. These systems are not perfect and may produce false positives or negatives. Our overall approach to moderation, reporting, enforcement, and appeals is described in the Trust & Safety Policy.
Your responsibilities as a user
- Use generated content lawfully and in accordance with the Terms of Service, Acceptable Use Policy, and Licensing & Usage Rights.
- Obtain any consents, licenses, and clearances needed for your intended use, especially for commercial release or public distribution.
- Review outputs for similarity, accuracy, and suitability before relying on them.
- Do not remove or defeat AI-generated content labels, and do not present synthetic media deceptively.
- Do not clone or synthesize a real person's voice or likeness without their explicit consent or another lawful basis.
Contact
Questions about this statement may be directed to legal@nestbitt.com or by mail to Nestbitt Inc., Suite C, 16644 - 71 St, Edmonton, AB. T5Z 0N5, Canada. This statement is governed by the laws of the Province of Alberta, Canada and will be updated as our AI systems and applicable transparency obligations evolve.