<\/p>\n
The UK government has paused its plans for broad copyright exceptions to AI training, opting for further evidence gathering and monitoring, diverging from the EU’s more interventionist approach and sparking debate over protecting creative industries and ownership of AI outputs.<\/p>\n<\/div>\n
The UK has stepped back from a previously favoured route for reforming copyright law to accommodate the training of artificial intelligence, opting instead for further evidence gathering and continued monitoring while industry and legislators on the Continent pursue a more interventionist path.<\/p>\n
According to the UK government\u2019s consultation materials and its subsequent statement of progress, ministers now say they do not view \u201ca broad copyright exception with opt-out\u201d as their \u201cpreferred way forward\u201d. The consultation, run jointly by the Department for Science, Innovation and Technology, the Intellectual Property Office and the Department for Culture, Media and Sport, set out control, access and transparency as its guiding objectives but left open a \u201cdo nothing\u201d option alongside three possible intervention models. The government\u2019s interim update in December 2025 signalled that a full report and economic impact assessment would follow.<\/p>\n
The choice to pause on immediate legislative change contrasts with regulatory developments in the EU, where the bloc\u2019s framework already provides for a wider text and data mining (TDM) carve\u2011out that permits AI training unless a rights holder expressly opts out in a machine\u2011readable form. Alexander Bibi of Pinsent Masons observed that \u201cEU law provides a broadly applicable text and data mining exception that enables AI training unless the rights holder explicitly exercises an opt\u2011out in a machine\u2011readable format,\u201d adding that many major rights holders in the EU have already built licensing offers aimed at AI training customers. The EU position rests on a combination of the 2001 information society directive, the 2019 digital single market directive\u2019s TDM rules and the 2024 AI Act\u2019s transparency duties for providers of general\u2011purpose AI.<\/p>\n
The European Parliament\u2019s recent non\u2011binding resolution has intensified debate in Brussels, urging tougher rights\u2011holder controls, exploration of an EUIPO opt\u2011out mechanism, stronger transparency requirements and examination of remuneration for past uses of protected works in training data. The European Commission and the Intellectual Property Helpdesk have previously highlighted that the 2019 directive introduced mandatory TDM exceptions while still allowing an opt\u2011out by rightholders, a compromise that shaped the current EU landscape.<\/p>\n
Within the UK, legal experts and legislators offer a different emphasis. A House of Lords committee report published in March 2026 warned that the government must choose between two AI futures and urged protection for the creative industries , which the report noted generated \u00a3124 billion in 2023 and employed 2.4 million people , favouring licensing\u2011based solutions to safeguard creators\u2019 incomes and cultural capacity. Pinsent Masons commentators and peers have argued that the UK must balance competitiveness against the need to preserve its creative sector.<\/p>\n
Beyond training datasets, Westminster has signalled a shift on ownership of AI outputs. The government said it will continue to monitor the use and impact of the protections afforded to computer\u2011generated works under the 1988 Copyright, Designs and Patents Act but proposed removing those protections \u201cin the absence of evidence of its ongoing value\u201d. Pinsent Masons noted that aligning protection to \u201chuman involvement\u201d mirrors the European droit d\u2019auteur approach and raises practical questions about how much human contribution is required to meet a creativity threshold and how such input would be proven in disputes. Industry analysis of TDM exceptions in the EU shows that the opt\u2011out mechanism has already influenced how rights holders and licensors design commercial offers for AI developers.<\/p>\n
Previous UK moves to expand data\u2011mining rights for commercial use have illustrated the policy tensions at play: past reforms aimed at permitting computational analysis where lawful access exists were intended to lower barriers to innovation but sparked concern among content owners about loss of control and revenue streams. That debate will remain central as ministers gather further evidence and stakeholders press for either clearer licensing routes or broader exceptions.<\/p>\n
Inspired by headline at:<\/strong> [1]<\/a><\/sup><\/p>\n