/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

Docs: Adobe offers its network of photographers $120 for videos of people engaged in everyday actions, or ~$2.62/minute to ~$7.25/minute on average, to train AI

- OpenAI's Sora has increased competitive pressure on Adobe  — Adobe has built AI models with stock library, procured media

Bloomberg Brody Ford

Context & Ripple Effects

Adobe's cautious effort to add generative AI to Creative Cloud had already been shaped by competition from DALL-E and Midjourney. The company had also built models from stock-library assets and procured media, making its contributor network a practical route to more targeted video material.

The request comes as reported AI-data deals were putting explicit prices on images and video. It also precedes scrutiny of AI-generated images in Adobe Stock used for Firefly training, underscoring how the provenance of training inputs is becoming part of product and creator relations.

First-order effects

  • Adobe creates a paid channel for photographers to supply action-focused video training data, while contributors gain a defined—if modest—new licensing opportunity.
  • The move gives Adobe a more purpose-built source of footage for its AI work rather than relying solely on its existing stock library and procured media.

Second-order effects

  • Other model builders seeking video capabilities face added pressure to secure licensed, task-specific footage; reported market rates for AI-data deals provide a visible benchmark for suppliers and buyers.
  • Adobe's contributors may expect clearer compensation and usage terms as their work shifts from conventional stock licensing to model training, especially after questions around Firefly's AI-generated stock inputs.

Third-order effects

  • Training data is becoming a separately priced production input, not simply a byproduct of stock-media distribution; platforms with contributor networks can turn those networks into a model-development asset.
  • If demand for licensed video data persists, competitive differentiation may depend as much on rights management and contributor trust as on the models themselves—though the reported payment levels do not establish an industry-wide standard.

The trend: Generative-video competition is pushing creative-software companies to formalize access to licensed, purpose-built training data.

Discussion

  • @adamhlavac @adamhlavac on x
    And yet, I STILL can't export a video without audio dropping out halfway through the video.
  • @klendathucap @klendathucap on x
    I can make a quick $300 with my WoW vods.
  • @brodyford_ Brody Ford on x
    Scoop: Adobe is paying around $3/minute for videos to build a new generative AI model. https://www.bloomberg.com/...