Inaccurate or Fabricated Content: RELX's 2025 Annual Report on the Risks of Generative AI, February 2026

RELX describes itself, in its Annual Report 2025, as "a global provider of professional information solutions." Its Strategic Report, which holds the section on principal and emerging risks, "has been approved by the Board," the report says, in a statement dated February 11, 2026. Parts of that section follow, not the whole. This article describes none of the company's products and reports none of its results.

How the Company Says It Weighs Risk

RELX says it has "a well-embedded risk management framework" based on the internal control framework of the Committee of Sponsoring Organizations of the Treadway Commission, and that it uses "the 3 Lines of Defence model." Its business areas, it says, are required to maintain systems of risk management and internal control appropriate to the nature and scale of their activities, and its Board monitors that system and assesses its effectiveness each year.

"Our risk management process considers the likelihood and impact of risks, the timeline over which a risk could arise, the direction in which risks are trending and the effectiveness of our mitigation efforts," the report says. Beyond current risks, the company says it also identifies "emerging risks which could impact our business in the next 3-5 years."

Inaccurate or Fabricated Content

"An emerging risk specific to generative AI is the possibility that models may produce inaccurate or fabricated content," the report says. The company says it reduces that exposure "by involving domain experts in the development, implementing comprehensive testing and validation procedures, and including verifiable, source-linked citations in AI-generated outputs to ensure users can access authoritative material."

Another emerging AI risk, the report says, lies in the journal publishing sector, where "Generative AI is lowering the cost and effort to produce fraudulent articles, either by innocent/ignorant or fraudulent author, editor or peer reviewer." The company says it combats this "with technological tools designed to identify fraudulently submitted articles."

Some of the Other Risks

The report mentions AI in other principal risks too. On customer demand, it says the means of delivery, customer demand and the products and services themselves "continue to change in response to technological innovations, such as the use of artificial intelligence," among other factors. On intellectual property, it says there is a risk that its proprietary rights and copyright protections "could be infringed or circumvented, including by companies leveraging technology tools and AI, which may impact demand for and pricing of our products and services," and that its contracts with customers contain provisions on the use of proprietary content "including use by large language models."

On personal data, the report says: "Our commitment to fair, explainable, and accountable AI practices, as set out in our Responsible Artificial Intelligence Principles, helps to ensure that our AI uses of personal data are subject to robust privacy governance." And on ethics, it says the company, its employees, major suppliers and partners are expected to adhere to high standards of integrity and ethical conduct, including those related to the "use of artificial intelligence."