NextFin News - On January 13, 2026, Roxanne Carter, Google’s head of intellectual property and public affairs, testified before the UK House of Lords Communications and Digital Committee in London. Carter articulated Google’s position that the company does not believe it should pay publishers or rightsholders for the use of most freely accessible online content in training its artificial intelligence (AI) models. This statement was made amid growing scrutiny over how AI companies utilize copyrighted materials for training large language models and generative AI systems.
Carter explained that AI training involves analyzing vast datasets to identify linguistic patterns and statistical relationships rather than copying or replicating content. She emphasized that Google’s AI products, such as AI Overviews and AI Mode, aim to generate wholly new content rather than reproduce publishers’ original works. While Google licenses content for extended display rights and specialized datasets, it maintains that licensing is not necessary for training on open web content.
During the hearing, Baroness Keeley, chair of the committee, pressed Carter on whether rightsholders should receive payment or revenue shares when their work is used by AI systems. Carter responded that if the content is freely available on the web, Google does not believe a license or payment is required. However, for off-platform or archived content, Google is engaging in licensing agreements. Carter also highlighted that publishers can opt out of AI training without being penalized in Google search rankings, referencing Google’s 2023 launch of “Google Extended,” a robots.txt-based control allowing content owners to permit search indexing but deny AI training access.
However, Carter declined to clarify whether publishers can effectively opt out of having their content scraped for Google’s AI Overviews, a product that summarizes news content and has been criticized for reducing publishers’ clickthrough rates by up to 50%. She noted ongoing discussions with the UK Competition and Markets Authority (CMA), which recently designated Google’s search and AI products as having strategic market status and is considering regulatory measures.
This testimony comes against a backdrop of increasing tension between AI developers and content creators worldwide. In December 2025, Google announced deals with major publishers including The Guardian, Der Spiegel, and The Washington Post, paying for extended display rights and API content delivery, though these were not described as licensing agreements for AI training data. Meanwhile, smaller publishers and creators have expressed concerns about the lack of compensation and control over their content’s use in AI training.
Google’s stance reflects a broader industry argument that AI training on publicly available data constitutes fair use or falls outside traditional copyright frameworks because AI models learn patterns rather than store or reproduce exact content. This position is contested by many publishers and rights organizations advocating for new licensing models and revenue sharing to sustain journalism and creative industries amid AI disruption.
From a strategic perspective, Google’s refusal to pay for most AI training content usage aims to minimize operational costs and maintain competitive advantage in the rapidly evolving AI market. The company’s willingness to license specialized or archived content indicates recognition of value in curated datasets but draws a clear line against broad licensing fees for open web content. This approach may influence global regulatory debates, as jurisdictions like the UK and EU explore frameworks to balance innovation incentives with fair remuneration for content creators.
Looking ahead, the ongoing dialogue between Google, regulators, and publishers will likely shape the future of AI training data governance. The CMA’s forthcoming rules could impose new obligations on Google and other AI firms, potentially mandating compensation or opt-out mechanisms with enforceable rights. Additionally, emerging markets such as India are proposing pioneering AI data licensing frameworks requiring royalties for copyrighted content use, signaling a trend toward stricter data sovereignty and creator compensation policies worldwide.
For content creators and publishers, the challenge remains to negotiate equitable terms that recognize their contributions without stifling AI innovation. For AI companies, clarifying licensing boundaries and investing in transparent, fair partnerships will be critical to sustaining growth and public trust. As AI-generated content increasingly competes with original journalism, the economic models underpinning digital media will undergo significant transformation, demanding adaptive strategies from all stakeholders.
In conclusion, Google’s IP head Roxanne Carter’s testimony underscores the complex intersection of AI technology, intellectual property rights, and market regulation. The company’s position against paying for most AI training content usage highlights the need for nuanced policy frameworks that reconcile technological progress with the economic sustainability of creative industries in the digital age.
Explore more exclusive insights at nextfin.ai.
