Training-Data Licensing and Content Provenance for Generative Media
Building image, video and avatar generators on licensed data and marking their outputs with provenance metadata or watermarks.
Tier Grades
Tier A: Frontier AI labs
feasible
Firms such as Adobe with 300 million licensed stock assets can train licensed-only models.
Tier B: Indonesian AI labs
stretch
An Indonesian lab rarely owns licensed media at that scale.
Tier C: Institution units such as ITB
feasible
Provenance and watermark studies fit a university unit.
Tier D: Organisations such as SEMAI (PoT, PoC, PoV)
feasible
Choosing licensed models and C2PA labelling is part of every client generative media pilot.
Where this comes from
Tiers: a) frontier AI labs, b) Indonesian AI labs, c) institution units such as ITB, d) organisations such as SEMAI delivering proof of technology, concept and value. Each rationale cites the anchors below.
Anchors
- ComputeTraining compute is not in the sources read.
- DataAdobe Firefly is trained on 300 million images and videos in Adobe Stock plus public domain and openly licensed sources.
- Team size (proxy)Proxy only (author count of an exemplar paper, not a headcount): not in the sources read.
- Time horizonYear: tied to court cases and AI law.
- Shelf-life signalsCourt rulings on training data, C2PA or watermark mandates in AI law
- Pipeline roleTier A and stock owners train licensed models; Tier C studies provenance; Tier D chooses compliant models and labels outputs in pilots.
Where this comes from
Team size is a proxy (author counts of exemplar papers), not a headcount.
Who and What
Exemplars
Observed actors
Linked models
Bria FIBOimage generationexemplar
Firefly Image 5image generationexemplarBria image modelsimage generationexemplar
Kling 3.0video generationproduced by
Seedance 2.0video generationproduced by
Nano Banana Pro (Gemini 3 Pro Image)image generationproduced byFirefly Image 4image generationproduced byRMBGvisionproduced byFirefly Imageimage generationproduced by
HeyGen Avatar IVavatarproduced by
Stable Diffusion 3.5image generationproduced byBria 3.2image generationproduced byLinks
R&D TopicTraining-Data Licensing and Content Provenance for Generative Media
ModelsFirefly Image 5Adobe Inc., image generationBria image modelsBria AI, image generationBria FIBOBria AI, image generationFirefly ImageAdobe Inc., image generation+8 more
Firefly Image 5Adobe Inc., image generationBria image modelsBria AI, image generationBria FIBOBria AI, image generationFirefly ImageAdobe Inc., image generation+8 moreUse Casesnone recorded
Who Runs Itnone recorded
Models 12
Kling 3.0Kuaishou, video generationproduced bySeedance 2.0ByteDance, video generationproduced byBria image modelsBria AI, image generationexemplarFirefly ImageAdobe Inc., image generationproduced byFirefly Image 4Adobe Inc., image generationproduced byFirefly Image 5Adobe Inc., image generationexemplarNano Banana Pro (Gemini 3 Pro Image)Google, image generationproduced byBria 3.2Bria AI, image generationproduced byBria FIBOBria AI, image generationexemplarStable Diffusion 3.5Stability AI, image generationproduced byHeyGen Avatar IVHeyGen, avatarproduced byRMBGBria AI, visionproduced byPapers
- Extracting Training Data from Diffusion ModelsNicholas Carlini et al., arXiv preprint, 2023 founding
- The Stable Signature: Rooting Watermarks in Latent Diffusion ModelsPierre Fernandez et al., ICCV 2023 (per arXiv comment), 2023 state of the art
- Tree-Ring Watermarks: Fingerprints for Diffusion Images that are Invisible and RobustYuxin Wen et al., arXiv preprint, 2023 state of the art
- Glaze: Protecting Artists from Style Mimicry by Text-to-Image ModelsShawn Shan et al., arXiv preprint, 2023 state of the art
- Foundation Models and Fair UsePeter Henderson et al., arXiv preprint, 2023 survey
Where this comes from
The papers to read first: where the direction began, its current best, and a survey when one exists.
Sources
- Adobe Firefly - Wikipedia en.wikipedia.org https://en.wikipedia.org/wiki/Adobe_Firefly
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Training-Data Licensing and Content Provenance for Generative Media", https://aurion.semai.id/topics/generative-media-licensing-and-provenance/, records of October 2026.