Government report sets aside AI training exception with opt-out
The UK government's report on copyright and AI says a broad exception with an opt-out is no longer its preferred approach.
On 18 March 2026 the government published its Report on Copyright and Artificial Intelligence, with an economic impact assessment, as required by sections 135 and 136 of the Data (Use and Access) Act 2025. It was published by the Department for Science, Innovation and Technology, the Department for Culture, Media and Sport and the Intellectual Property Office.
The report states that the 2024 to 2025 consultation received 11,520 responses. It says the originally preferred proposal, a broad text and data mining exception with an opt-out, was rejected by most respondents, and that 81% of respondents chose the option of strengthening copyright so that licensing is required for AI development.
The government concludes that a broad exception with opt-out is no longer its preferred way forward. It proposes to gather further evidence on how copyright law affects the development and use of AI, to consider other policy approaches, and to monitor technology, litigation, international approaches and the licensing market. The report also covers input and output transparency, technical standards, licensing, enforcement, computer-generated works and digital replicas.
Why it matters: for now, UK copyright law is unchanged and there is no new exception for AI training. Licensing remains the main route by which authors and publishers can be paid for such use, and the report leaves the questions of transparency and enforcement open for further work.
Sources: GOV.UK: Report on Copyright and Artificial Intelligence; GOV.UK: Copyright and artificial intelligence (consultation page).