AI licensing deals give big publishers a ChatGPT edge

By mid-2026, publishers have increasingly capitalized on their brand authority and editorial trust as valuable assets in the AI landscape, with 51% of 110 U.S. publishers having signed AI or LLM licensing deals and another 35% negotiating, according to an IAB report. This momentum is driven by users’ skepticism toward AI outputs—60% of AI users say a company’s reputation influences their trust in AI responses—creating a lucrative opportunity for publishers to monetize their credibility. However, this licensing uptake is uneven, with 60% of large publishers securing deals compared to only 20% of smaller independents, underscoring resource disparities in negotiating with AI firms and shaping a new ecosystem where trusted content fuels AI discovery, traffic, and fair compensation, as Jack Koch of the IAB describes it.

Getty Images’ pioneering partnership with OpenAI exemplifies the evolving licensing models that balance revenue generation with content protection, introducing a novel ‘display-only’ agreement that permits licensed content to appear in ChatGPT search results without being used for AI training. This approach, coupled with Getty’s own ‘commercially safe’ AI Generator trained exclusively on licensed data, marks a strategic departure from competitors relying on web-scraped content and addresses mounting legal and ethical concerns. The proliferation of similar deals—such as Shutterstock’s partnerships with Nvidia, Meta, Amazon, and OpenAI—signals a broader industry shift toward formalized, recurring revenue streams in the AI era, though smaller creators often remain excluded and undercompensated, as highlighted by UK visual artists’ widespread lack of permission and awareness regarding AI training use.

Recent research by Press Ranger and OtterlyAI reveals that publishers with OpenAI licensing agreements enjoy a substantial citation advantage on ChatGPT, garnering 48% more AI citations and 112% more citations per page than unlicensed peers, with licensed publishers’ content drawing 57.9% of their AI citations from ChatGPT alone. Yet, this citation premium appears platform-specific; Google-licensed publishers do not see similar benefits on Google AI Overviews, and Perplexity’s licensed publishers perform at parity with unlicensed ones. Moreover, licensing is not the sole path to AI visibility—unlicensed outlets like NerdWallet and Healthline remain highly cited, and niche and trade publishers, mostly unlicensed, dominate AI citations across 15 of 16 U.S. industries, suggesting that strategic content formats and targeted niches remain critical alongside licensing deals.

Content formats favored by AI platforms tend to be evergreen and commercially oriented, with best-of lists, buying guides, and product reviews comprising nearly 47% of citations for licensed publishers. This insight underscores the importance for publishers and PR teams to complement licensing agreements with strategic content creation that aligns with AI citation patterns, thereby maximizing both visibility and compensation opportunities in an increasingly complex publisher-AI partnership ecosystem.

The legal battleground over AI content use is epitomized by Getty Images’ lawsuit against Stability AI, which accused the latter of unauthorized training on millions of copyrighted images. Although the High Court in London dismissed most claims, the ongoing appeal underscores the unresolved tension between content ownership and AI development. Meanwhile, strategic licensing agreements, such as Getty’s 'display-only' deal with OpenAI, are emerging as a novel model that permits AI firms to showcase licensed content in search results without using it for training, thereby safeguarding publishers’ rights and revenue streams. This trend is mirrored across the industry, with major press publishers like Associated Press and Reuters, as well as image libraries like Shutterstock, forging partnerships with AI developers including Nvidia, Meta, Amazon, and OpenAI to establish authorized, attributed content as a valuable asset in AI services.

Regulatory frameworks are rapidly evolving to address the complexities of AI content use, with the EU AI Act 2024 and UK copyright policy shifts imposing transparency and consent obligations on AI developers. These legal changes empower publishers by demanding clarity on how AI models are trained and curbing unauthorized data usage. Concurrently, publishers face mounting challenges from AI-driven search engines that have caused a 33% year-over-year referral traffic loss and led to 68% of Google searches ending without clicks. In response, publishers have launched lawsuits—such as 400 newspapers suing OpenAI and Microsoft—and regulatory bodies like the UK Competition and Markets Authority have mandated greater transparency and control over AI content usage, including opt-out options and impression-based reporting, signaling a new era of publisher leverage and accountability for AI misinformation, as seen in the Munich court ruling holding Google liable for AI-generated falsehoods.

Major media companies are aggressively defending their intellectual property through a mix of litigation and licensing, exemplified by News Corp suing Brave to protect its Microsoft Copilot deal and CNN’s lawsuit against Perplexity AI after failed negotiations. This carrot-and-stick approach—where publishers license content to some AI firms while threatening or pursuing legal action against others—reflects a strategic effort to establish a robust licensing market that can demonstrate market harm in copyright disputes. The Department of Justice’s antitrust case against Google, seeking divestiture of its AdX ad exchange for monopolistic practices, further complicates the landscape by potentially reshaping digital advertising and AI data ecosystems. Meanwhile, alliances like the News/Media Alliance are pioneering licensing deals that enable smaller publishers to tap into the $2 billion Retrieval-Augmented Generation market, although skepticism remains about whether these arrangements will translate into meaningful revenue beyond PR.

Despite an estimated 80 to 100 ongoing lawsuits in the US alone, none have yet proceeded to trial, with recent high-profile settlements such as Anthropic’s $1.5 billion payout to authors signaling a growing preference for negotiated resolutions. However, the legal landscape remains unsettled, as cases approach summary judgment phases that could set influential precedents. The novelty and complexity of AI content licensing are further complicated by AI companies’ nuanced valuation of publisher content, where entities like OpenAI weigh the influence or prestige of sources differently—questioning, for instance, whether the New York Times’ content holds more weight than other publishers. Resources like Columbia University’s Tow Center are critical in tracking these evolving deals and disputes, highlighting the ongoing flux and the high stakes for publishers seeking fair compensation and control in the AI era.

By mid-2026, it became clear that AI platforms such as ChatGPT, Perplexity, Google AI Overview, and Grok cite overwhelmingly distinct sources for identical queries, with studies showing that up to 88% of cited pages appear uniquely on only one platform. This divergence means that publisher visibility is highly platform-specific, debunking the common misconception that all large language models operate under a uniform citation ecosystem. Consequently, brands must tailor their AI strategies to each engine’s unique source preferences and trust signals, as failing to appear on a given platform effectively excludes them from that AI-driven conversation.

Licensing deals, particularly with OpenAI, confer a significant citation advantage on ChatGPT, with licensed publishers earning 48% more citations and 112% more citations per page compared to unlicensed peers. However, this boost is platform-specific; Google AI Overviews and Perplexity show no comparable premium for their licensed publishers, underscoring that licensing tilts visibility heavily toward ChatGPT but does not guarantee broader AI presence. Notably, some unlicensed publishers like NerdWallet and Healthline still rank highly, indicating that licensing is a powerful but not exclusive path to prominence.

The citation landscape also varies in volume and source type across AI platforms, with Grok citing an average of 28.3 sources per response—over four times more than ChatGPT’s 6.2—offering more citation opportunities but also intensifying competition among agencies. Established directories like Clutch.co dominate citations across multiple AI models, appearing in the top five domains for Grok, Microsoft Copilot, and Google AI Overview, highlighting the enduring value of licensing deals and platform prominence. Simultaneously, there is a rising trend of AI models increasingly citing individual agency websites, such as Made Agency and Folklore, signaling that agencies must optimize their own sites alongside leveraging directories to maximize AI visibility.

AI platforms like ChatGPT and Perplexity employ distinct trust signals and URL prioritization methods, resulting in minimal overlap in cited sources and creating a fragmented landscape for brand visibility. This divergence necessitates that publishers and marketers adopt multi-platform monitoring tools such as Semrush's AI visibility score to track and optimize their presence across diverse AI ecosystems, ensuring data-driven adjustments to PR and SEO strategies that cater to the unique citation behaviors of each AI system.

Data-led press releases that foreground specific, economically relevant numbers dramatically increase AI citations—earning 3.5 times more citations according to LeadCoverage CEO Kara Brown, who emphasizes that 'AI cannot invent a number, so it cites whoever published one.' This strategic use of concrete data not only boosts AI-driven visibility but also translates into an 83% rise in search impressions, even as traditional search clicks decline, reflecting a paradigm shift where AI-generated answers become the primary buyer touchpoint.

Holding an OpenAI licensing deal confers a significant advantage on ChatGPT, with licensed publishers earning 48% more citations and averaging 10.2 citations per page versus 6.9 for unlicensed peers, as revealed by a comprehensive study from Press Ranger and OtterlyAI analyzing 129.3 million citations. However, licensing is not a silver bullet; unlicensed niche and trade outlets like NerdWallet and Healthline often outperform mainstream media in many industries, underscoring the importance of a multi-faceted, data-driven strategy that includes pitching AI-favored content formats such as best-of lists and buying guides to maximize AI engagement.

The study by Press Ranger and OtterlyAI also highlights that while OpenAI licensing deals provide a strong home-platform advantage on ChatGPT—accounting for 57.9% of licensed publishers’ AI citations—this advantage does not necessarily extend to other AI platforms like Google AI Overviews or Perplexity, where licensed publishers often see parity or lower citation rates compared to unlicensed competitors. This finding reinforces the need for publishers and PR teams to leverage comprehensive, multi-platform AI citation data and interactive dashboards to tailor content strategies that optimize visibility across the entire AI landscape.

Smaller and mid-sized publishers face steep hurdles in securing AI licensing deals compared to their larger counterparts, largely due to limited resources and bargaining power. By mid-2026, only 20% of smaller independents had signed AI or LLM licensing agreements versus 60% of large publishers, highlighting a significant disparity. This challenge is compounded by uncertainty around pricing mechanisms and verification of their contributions to AI outputs, raising existential questions about their role and influence in the emerging content-for-training market.

In response to these challenges, industry initiatives have emerged to support smaller publishers’ navigation of AI licensing and compensation complexities. Programs from the Professional Publishers Association and the News/Media Alliance, such as the latter’s comprehensive AI licensing deal centered on recurring Retrieval-Augmented Generation (RAG) revenue, offer crucial lifelines amid declining ad revenues and traffic. However, skepticism persists about whether these agreements will yield meaningful compensation, given the historical precedent of AI models using publisher IP without payment or legal obligation.

Efforts to enhance transparency and control for smaller publishers are also gaining regulatory backing, exemplified by the UK Competition and Markets Authority’s mandate requiring Google to provide opt-outs and clearer attribution in AI search results. Such measures could empower smaller publishers to better protect their content and negotiate fairer terms. Complementing these efforts, academic resources like Columbia University’s Tow Center maintain a detailed ledger of AI deals and lawsuits, effectively serving as an informal industry ombudsman that helps smaller publishers stay informed and advocate for their rights in the evolving AI landscape.