Your pipeline is professional. The industry around it has no guardrails.
Studios run controlled, contracted pipelines. The relationships with talent work. But the industry has no shared infrastructure to prove it — and no way to stop what happens outside those relationships. ByteDance’s Seedance generated hyper-realistic Tom Cruise and Brad Pitt footage from public data alone. Disney, Paramount, Warner Bros., Netflix, and Sony all sent cease-and-desist letters within days — but had no technical mechanism to prevent it, prove their own authorizations, or verify scope at the pipeline level. Meanwhile, synthetic replacements get trained on public footage by entities that never hired anyone, and post-acquisition strangers inherit performance data with no relationship to the talent. DIAP is the guardrail system that the professional pipeline needs: shared infrastructure for studios, agencies, guilds, and talent to control authorization, prove provenance, and protect everyone’s IP.
The Performance Signature. Motion, timing, physicality — the data AI learns from.
$28B+
AI in media & entertainment (2025)
160K+
SAG-AFTRA members affected
0
Industry-wide authorization standards
What DIAP Does For Studios
Project-Scoped License Tokens — Your IP Shield
Every token is locked to a specific project, production, or campaign. A token for Film A cannot be reused for Film B. Scope, duration, rights, territory, and render limits are baked into the token itself — cryptographically. This protects the studio and the talent from the same threats: if a downstream vendor, co-production partner, or post-acquisition entity tries to reuse identity data outside scope, the token proves where the boundary was. It’s infrastructure that travels with the asset.
Two-Layer Authorization — Visibility Before Usage
Layer 1 controls whether a talent’s identity is visible to your pipeline. Layer 2 controls per-project usage. Token requests are routed through the talent’s representative or vault — mirroring how talent representation already works. You get introduced first, then you negotiate. DIAP makes the introduction cryptographic.
Agent & Representative Access
Most working talent is managed by agents, managers, or attorneys. DIAP’s delegation model mirrors the real industry: agents can manage consent on behalf of their roster — approving token requests, setting visibility preferences, reviewing audit logs. Studios interact with agents the same way they always have, but now the consent trail is cryptographic, not contractual. Agents can build branded portals (like a "CAA Identity Manager") on top of DIAP’s API to manage their entire roster’s consent in one interface.
TRAINING_USE — Always a Separate Right
Using a talent’s face to render a poster is not the same as using it to train a model. DIAP verifies that distinction at the protocol level. TRAINING_USE must be explicitly granted and is never implied by any other right. This applies to identity data and to scripts. This protects studios too: if a downstream AI vendor claims "implied training rights" from a render token, the studio can prove the authorization was scoped to render only.
ScriptModule — IP Protection for Written Works
Scripts, story bibles, treatments, and dialogue are protected assets. ScriptModule gives studios a cryptographic provenance trail for every script accessed in their AI pipelines — who accessed it, when, under what authorization, and whether TRAINING_USE was granted. AuthorshipReceipts create an unbroken chain of custody that protects both studio and writer. If a script leaks to an unauthorized party or surfaces in a model trained overseas, the provenance chain shows exactly where authorization ended.
Render Receipts — Per Frame
Every frame rendered using a talent’s identity generates a cryptographic receipt. Receipts link back to the authorizing token, creating an unbroken provenance chain from consent to output. This is the infrastructure that lets studios, talent, and unions all see the same truth: exactly what was authorized, what was rendered, and when. If an output surfaces somewhere it shouldn’t — a foreign platform, an unauthorized derivative — the receipt chain shows where it came from and where authorization ended.
Dual-Layer Watermarking
Every authorized output carries two independent watermarks: a pixel-level watermark payload (invisible to the eye) and an ultrasonic audio frequency fingerprint (inaudible) with a unique Sound ID. Both layers survive compression, re-encoding, and social media upload. If a studio’s content appears on an unauthorized platform, the watermark proves it was theirs — and the render receipt proves it was authorized.
Revocation Response — Minutes, Not Weeks
If authorization changes — a talent revokes, a token expires, terms are updated — DIAP’s heartbeat system flags all active render pipelines immediately. The studio knows within minutes, not weeks. In-progress renders are flagged before distribution. This is infrastructure that protects the relationship: fast notification means fast resolution, before anything leaks outside the studio’s control.
How Studios Implement DIAP
DIAP integrates into existing production infrastructure through a REST API and SDKs. The typical integration: your production management system (ShotGrid, ftrack, or custom) calls DIAP’s API to request a license token before any AI render job starts. The VFX pipeline checks the token on every render call. Receipts are generated automatically. No new software to install — DIAP lives as an API layer between your production management and your render farm. For Unreal Engine, Unity, or custom Python pipelines, DIAP provides direct SDK integration that hooks into your existing asset pipeline.
Build Your Own Layer on DIAP
Studios can build their own branded tools on top of DIAP’s open API — a casting consent dashboard, a compliance tracker for VFX supervisors, a distribution clearance tool for legal. Your tools, your interface, your workflow. DIAP provides the authorization infrastructure, the token engine, and the audit trail. You build whatever your production needs on top of it.
How Studios Integrate DIAP
Connect your pipeline to DIAP’s API
Register as a studio on DIAP and get API credentials. Connect your production management system (ShotGrid, ftrack, or custom) to DIAP’s REST API. When a VFX supervisor starts a shot using a talent’s digital likeness, the system calls DIAP to request a license token. No new software — it’s an API call in your existing workflow.
Talent or agent approves with terms
The token request goes to the talent’s Identity Vault — or to their agent’s management dashboard if authorization is delegated. The agent reviews, approves with specific conditions, and the token is issued with all terms cryptographically embedded. Studios interact with agents the same way they always have.
Render with automatic compliance
Your render pipeline checks the token on every call. Each frame generates a cryptographic receipt automatically. Every output is watermarked with dual-layer provenance (pixel + audio). The studio’s compliance record builds itself as the production runs — no manual logging required.
Export provenance for compliance and distribution
At any point, export compliance bundles: token histories, render receipts, AuthorshipReceipts, watermark verification reports. These are machine-readable provenance records for union reporting, distribution clearance, E&O insurance, and regulatory requirements.
VFX team renders 2,847 frames. Token expires. DIAP flags it before distribution.
VFX team requests FACE_RENDER token for 90-day production window
2,847 frames rendered — each generates a cryptographic render receipt
Token expires at day 90 — heartbeat system detects loss of authorization
All active render pipelines flagged — distribution blocked automatically
Studio requests token renewal — talent approves with updated terms
What Studios Get
- Project-scoped license tokens — cryptographic IP boundaries per production
- Two-layer consent — visibility before usage, mirrors real talent representation
- Agent & representative delegation — work with talent reps the way you already do
- TRAINING_USE always a separate, explicit right — protects studios from downstream claims
- ScriptModule — IP protection and AuthorshipReceipts for written works
- Per-frame render receipts — unbroken provenance from consent to output
- Dual-layer watermarking (pixel + ultrasonic Sound ID) on all outputs
- Instant revocation alerts — limits studio exposure to minutes, not months
- REST API + SDK integration into ShotGrid, ftrack, Unreal, Unity, or custom pipelines
- Build branded tools on DIAP — casting dashboards, compliance trackers, clearance tools
- Provenance export for unions, distribution, and compliance requirements