Content Licensing & Copyright Regime
Publisher licensing deals & EU/US copyright constraints on training
Training-data provenance became a first-order constraint by 2025: labs signed licensing deals with publishers (News Corp, Axel Springer, Reddit, Shutterstock) while facing litigation (NYT v OpenAI) and EU AI Act copyright-transparency duties plus the text-and-data-mining opt-out under the DSM Directive. This legal layer now gates what the tiktok-scale-data corpus can legally contain.
EU basis
DSM Directive TDM opt-out
Deals
Axel Springer, News Corp, Reddit
Litigation
NYT v OpenAI
How it fits the stack
Content Licensing & Copyright Regime with what it depends on (above) and what it feeds (below). The figure renders as a crawlable diagram and upgrades to an interactive 3D graph as it scrolls into view.
Content Licensing & Copyright Regime in the AI stack. Content Licensing & Copyright Regime with its immediate upstream dependencies (top) and downstream dependents (bottom) in the AI value chain. Hover a node in 3D, or read the full relationships below.
Graph data (text) — 4 entities, 3 relationships
- Content Licensing & Copyright Regime —depends on→ EU AI Act
- EU AI Act GPAI Code of Practice —depends on→ Content Licensing & Copyright Regime
- Training Data & Web Corpus —depends on→ Content Licensing & Copyright Regime