Answers are grounded in this paper and cite their sources.
Purpose This study examines strategic content licensing between a generative artificial intelligence (AI) platform and a digital news publisher in digital news markets, an important issue for understanding where AI platforms rely on publisher content while competing for user attention. It analyzes how citation intensity shapes licensing incentives, pricing, user diversion and welfare. Design/methodology/approach We develop a game-theoretic model of two-dimensional horizontal differentiation in which the AI platform has a comparative advantage in content integration and the publisher has a comparative advantage in factual accuracy. The model examines how citation intensity, hallucination risk, user preference correlation and per-use royalty jointly determine equilibrium outcomes. Findings Licensing between AI platforms and publishers arises from a trade-off between two countervailing forces: entry displacement and engagement expansion. Content licensing requires sufficiently high citation intensity; mutual gains additionally require low preference correlation. The pricing effects of citation intensity are nonmonotonic. Subscription prices can rise or fall depending on the strength of citation relative to royalty costs. User diversion is shaped by the balance between referral traffic expansion and direct substitution. Welfare improves only when engagement expansion dominates entry displacement. Originality/value This study models AI–publisher licensing in a two-dimensional Hotelling framework of asymmetric comparative advantages, conceptualizing citation intensity as a content-composition variable that shapes output quality, user traffic and royalty payments beyond existing display-level constructs.
No takes yet. Share an insight, caveat, or question.
Diao et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: