OpenAI licensing deals lift ChatGPT citations 48%, joint study finds

A study of 129.3 million AI citations found OpenAI-licensed publishers earn 48% more ChatGPT citations per page than unlicensed peers.

OpenAI licensing deals lift ChatGPT citations 48%, joint study finds

A joint study by PR distribution platform Press Ranger and AI search monitoring firm OtterlyAI has put a number on a question that has divided publishers and communications teams since OpenAI began signing content licensing deals: publishers with an OpenAI agreement earned 48% more citations per page on ChatGPT during June 2026 than those without one.

The study examined 129.3 million citations across seven AI search platforms, including ChatGPT, Google AI Overviews, Google AI Mode, Perplexity, Microsoft Copilot, Gemini, and Claude, and matched them against a database of 91 confirmed AI-publisher licensing agreements mapped to 314 domains. Pages from OpenAI-licensed publishers averaged 10.2 citations each on ChatGPT, versus 6.9 for unlicensed pages. Across all seven platforms combined, the premium narrowed slightly to 46%.

The platform concentration effect

The data reveals a stark concentration dynamic. Publishers that signed exclusively with OpenAI earned 112% more citations per page on ChatGPT than unlicensed publishers, well above their average premium on any other platform. As a consequence, OpenAI-licensed publishers now draw 57.9% of their total AI citation volume from ChatGPT alone, while unlicensed publishers maintain a more balanced spread across ChatGPT and Perplexity.

Notably, the home-platform advantage did not replicate for other licensors. Google-licensed publishers were cited on Google AI Overviews at a slightly lower rate than comparable unlicensed publishers, and Perplexity's licensed publishers landed at parity on their platform. The asymmetry suggests that OpenAI's deals may involve forms of preferential data access or retrieval weighting that Google's and Perplexity's equivalent arrangements do not currently replicate.

"A licensing deal does one clear thing: it tilts your citations toward ChatGPT," said Thomas Peham, chief executive of OtterlyAI. "It does not guarantee you show up everywhere, and it is not the only way to win."

Trade and niche outlets lead outside news

The study also found that news pages account for just 7.2% of all AI citations, and that licensed news citations are heavily concentrated among five media groups: Future plc, Forbes, People Inc., Condé Nast, and Hearst, which together capture 69% of all citations going to licensed publishers. Yet across 16 US industries examined, niche and trade outlets, the majority of them unlicensed, carried most news citations in 15 of the 16 categories and collected 213% more AI citations than mainstream media.

For communications and PR teams, the practical implication is that a licensing deal is neither necessary nor sufficient for broad AI visibility. The study found that commercial and evergreen content formats, including best-of lists, buying guides, and product reviews, account for 46.9% of licensed publishers' AI citations. The authors recommend that PR teams map which publishers already earn citations in their verticals and focus pitches on service-journalism formats that AI systems demonstrably favour.

Market and regulatory context

The AI citation economy is emerging as a significant strategic question for media owners and brand marketers simultaneously. Publishers argue that AI systems training and citing their content without compensation is a value transfer that licensing agreements begin to address; platforms counter that citations drive referral traffic. The resulting negotiation dynamic sits uneasily with existing copyright frameworks in both the UK and EU, where several collective licensing proposals are under active discussion.

For enterprise software and marketing technology vendors, the growth of generative engine optimisation (GEO) as a discipline sits alongside, and increasingly competes with, traditional search engine optimisation budgets. OtterlyAI, which contributed the citation dataset, reported more than 40,000 marketing professional users as of August 2026 and was named in Gartner's 2025 Cool Vendors in AI in Marketing report, indicating that the tooling market around AI search visibility is already consolidating around a small number of early movers.