Provenance Manifest

Per-asset disclosure for: Dental Implants: $48,000 in the US, $4,800 in Turkey. Worth It?

Last updated: 2026-07-17 · EU AI Act Article 50 (Aug 2 2026 enforcement) deployer disclosure compliant.

Why this page exists

Wellness Vision produces faceless cost-comparison documentary content using an AI-assisted production stack. Under the European Union AI Act Article 50 (effective 2 August 2026), deployers of generative AI systems must disclose synthetic media to viewers in clear, conspicuous, and accessible form.

This page enumerates every AI-generated or AI-assisted asset that contributed to the video above. It is the canonical provenance reference for editorial transparency, regulatory compliance, and audience trust.

Disclosure stack

Description footer (YouTube + cross-platform): AI Disclosure block listing all four production tools.
YouTube Studio toggles: Altered content + Synthetic content (Data API v3 containsSyntheticMedia: true).
Per-platform manual disclosure: TikTok AI Disclosure per-Short; Instagram Meta AI Content label per-Reel; X thread AI disclosure footer.
C2PA Content Credentials: DALL-E thumbnail credentials verified with c2patool at the release gate.
This transparency page: Per-asset IPTC digitalSourceType + SynthID + C2PA status enumerated below.

Per-Asset Manifest

Voice Narration

Microsoft Azure Neural TTS

Model: en-US-Emma2:DragonHDLatestNeural
Parameters: Prosody rate +13%, documentary narration style
IPTC digitalSourceType: trainedAlgorithmicMedia
SynthID watermark: Not present
C2PA Content Credentials: Not present
License: Azure Cognitive Services TOS — commercial broadcast permitted

AI-generated voice for the entire narration. No human voice actor employed.

Thumbnail

OpenAI DALL-E

Model: gpt-image-1.5
Parameters: HD, prompt locked to D-W057 brand palette
IPTC digitalSourceType: trainedAlgorithmicMedia
SynthID watermark: Not present
C2PA Content Credentials: Present
License: OpenAI Content Policy + Wellness Vision commercial deployment

AI-generated thumbnail. C2PA Content Credentials (OpenAI Media Service) present on the source render and verified with c2patool; the resized upload copy carries IPTC digitalSourceType=trainedAlgorithmicMedia. The resize is required by YouTube's 2MB thumbnail limit and, because C2PA signs the pixels, necessarily invalidates the signature on the copy — the signed source is the artifact to verify against.

B-roll Visual Clips

xAI Grok Imagine

Model: grok-imagine-v6
Parameters: Documentary aesthetic, geographic anchor disambiguated, brand-neutral clinic visuals
IPTC digitalSourceType: trainedAlgorithmicMedia
SynthID watermark: Not present
C2PA Content Credentials: Not present
License: xAI Premium commercial use permitted

All B-roll AI-generated. No stock footage, no real-clinic photography, no recognizable human faces (LEGAL §3.3). Grok emits no SynthID/C2PA — manual disclosure on every surface compensates (SYNTHID_GAP_DISCLOSURE_PLAYBOOK).

Background Music

Soundraw

Model: Soundraw AI Composer (Creator Plan)
Parameters: Per-segment filter spec; Audacity Auto Duck pre-mix under VO
IPTC digitalSourceType: trainedAlgorithmicMedia
SynthID watermark: Not present
C2PA Content Credentials: Not present
License: Soundraw Creator Plan — commercial broadcast permitted, royalty-free

AI-composed music. Soundraw selected post-Suno RIAA litigation (D-W009).

Captions

OpenAI Whisper

Model: whisper-large-v3
Parameters: Word-level transcription of the AI voiceover for on-screen captions + timing reconcile
IPTC digitalSourceType: humanEditedAlgorithmicAssistance
SynthID watermark: Not present
C2PA Content Credentials: Not present
License: OpenAI Whisper open-source MIT

Transcription of AI-generated voice — algorithmic transcription assistance, not synthetic-of-itself.

Honest Caveats

Prices are point-in-time editorial comparisons. All figures are sourced from the high-trust public sources listed in the video description and may change; nothing on this page or in the video is a quote or an offer.
Most assets lack SynthID watermarks. Grok Imagine, Azure Neural TTS (user-perceivable layer), and Soundraw do not emit SynthID at output. Wellness compensates with explicit human-readable disclosure on every surface (description footer + Studio toggles + per-platform manual + this transparency page).
Microsoft Speech Watermark is separate from SynthID. Azure Neural TTS embeds a Microsoft-proprietary speech watermark for traceability, but it is not detectable by SynthID-aware badge systems (SYNTHID_GAP_DISCLOSURE_PLAYBOOK §3).
EU AI Act Article 50 obligations remain with Wellness as deployer. Even where individual tools lack output watermarks, Wellness as the deploying creator bears the disclosure obligation in jurisdictions where viewers access the content.
No human voice actor was employed. The narration is fully AI-generated. Wellness does not impersonate real persons (Article 50 make-a-real-person-appear-to-say-something clause not triggered).

Regulatory Compliance Posture

Regulation	Wellness Compliance Surface	Status
EU AI Act Article 50 (Aug 2 2026)	AI Disclosure footer + Studio toggles + this transparency page + IPTC digitalSourceType per asset	Prepared — verified at publish
YouTube Altered/Synthetic Content Policy	Studio Altered + Synthetic toggles + Data API v3 `containsSyntheticMedia: true`	Prepared — verified at publish
FTC AI Disclosure Guidance	"Produced with AI assistance" in description body + this transparency manifest	Prepared — verified at publish

Editorial Policy

Wellness Vision does not book trips, does not receive clinic referrals, and does not recommend specific providers. All cost data derives from the high-trust public sources cited in the video description. This is a cost comparison, not medical advice.

For corrections, additional disclosures, or reproducibility questions related to this video's provenance, contact [email protected].

You shouldn't have to go broke to feel whole.