AI Content Licensing Market Mirrors Search Era's Publisher Squeeze

The emerging market for AI content licensing is replicating the structural problems that damaged journalism during the search and social media eras—only faster and with greater concentration of power, according to new research from the Brookings Institution.

A report titled "Same Gatekeepers, New Tollbooths: Mapping the AI Content Licensing Market" finds that publishers who signed direct licensing deals with AI companies initially saw better click-through rates from AI interfaces. By the fourth quarter of 2025, however, that advantage had vanished amid a six-fold collapse in AI-driven referrals across the board. Publishers without deals experienced worse absolute numbers but smaller proportional drops. Both groups lost ground.

The findings align with independent analyses from the U.K. House of Lords and European Parliament, which reached strikingly similar conclusions using different methodologies—suggesting a narrow window for policy intervention remains open.

Why it matters

The deal structures, pricing precedents, and governance norms being established now will prove difficult to reverse once normalized. AI systems depend on continuous supplies of high-quality human content to remain useful, giving publishers more leverage than they typically exercise. But the same Big Tech firms whose AI products erode website traffic now build and control the licensing infrastructure publishers must use for compensation—a dynamic that doesn't require bad intent to produce harmful outcomes.

Three-tier market fails most publishers

The licensing market operates across three distinct tiers, each with significant limitations.

The first tier consists of bilateral deals between major AI companies and select publishers with national or global brand recognition. These confidential agreements involve real money flowing to real newsrooms, but publishers negotiate without visibility into how their content is used, at what frequency, or to what commercial effect. Limited disclosure makes it difficult for rights-holders to enforce their rights.

The second tier comprises intermediaries—a field that expanded from a handful of Silicon Valley startups to more than a dozen companies since 2024. These venture-backed firms offer bot detection, content marketplaces with pay-per-use pricing, and attribution-based revenue distribution. While they provide meaningful improvements over opaque bilateral deals, most remain vulnerable to acquisition by the same large technology companies from which they nominally protect publishers.

The third tier—local newspapers, regional broadcasters, ethnic and indigenous media, non-English language publishers, and individual creators—remains effectively absent from the licensing market entirely. This represents a structural failure to value how journalism's civic contribution is actually distributed.

Valuation framework ignores AI's full dependence

Most licensing negotiations rest on a narrow conception of value: referral traffic lost. This framing ignores publishers' contributions across training and fine-tuning, linguistic capacity, factual grounding, temporal currency, and civic legitimacy.

Copyright discourse has largely treated content-scraping as a discrete historical event. But with retrieval-augmented generation, models trained on publisher content activate anew with every inference call. Publishers accepting current deal terms may be foreclosing substantially larger claims courts have not yet adjudicated.

Policy options exist but require political will

Several fixes remain achievable within existing legislative traditions: statutory licensing frameworks with set rates, collective licensing and sectoral bargaining, mandatory transparency on deal terms and data usage, attribution systems at the model inference layer, and explicit inclusion requirements for local and independent media.

Australia and Canada have demonstrated that bargaining code frameworks work legislatively. The music industry has shown collective licensing scales effectively. The architectural blueprints exist—what's missing is political will to apply them before market structures calcify.

The research was co-authored by scholars at the Center for Journalism and Liberty at the Open Markets Institute and first published by the Brookings Institution.