UK’s stance on generative AI & copyright — what businesses need to know

With generative AI transforming how content is created, distributed and monetised, tensions have been heightened between the UK’s world-leading creative industries and the rapidly expanding tech sector.

In March 2026, the House of Lords Communications and Digital Committee published its AI, copyright and the creative industries report, which was followed by a further House of Commons report on copyright and artificial intelligence alongside a written ministerial statement from the Secretary of State for Science, Innovation and Technology that set out the Government’s revised approach. Taken together, these developments mark a clear shift in the UK’s direction of travel on AI and copyright. 

Here, Paddy Fearnon, Natalie Hardman and Colin Bell breakdown the key findings, the Government’s current stance and the five practical steps that businesses should consider when navigating AI and copyright compliance in the UK.

UK parliamentary reports — why they matter

Both the House of Lords and House of Commons reports start from the same premise: the UK’s creative industries are a national economic asset, contributing well over £100bn in gross value added and supporting millions of jobs across the country. Copyright is described as central to that success, providing the legal framework that underpins investment, creativity and commercialisation.

At the same time, Parliament recognises AI as a strategic technology and pillar of the UK’s industrial strategy. The policy challenge isn’t whether the UK should support AI but how to do so without weakening the copyright framework that sustains the creative economy.

For businesses, this framing matters. It signals that future reform is unlikely to prioritise speed or deregulation at the expense of legal certainty. Instead, the emphasis is on balance, evidence and long-term market stability.

No new copyright exception for AI training data in the UK

A central point of alignment between the two reports is the rejection of a broad copyright exception permitting text and data mining for AI training.

The House of Lords was explicit that introducing a new exception — particularly one based on rights holders opting out — would undermine copyright protection without clear evidence that it would deliver meaningful economic benefits. The House of Commons report and the Government’s progress paper reinforce that position.

In her written statement to Parliament, the Secretary of State confirmed that the Government has stepped back from its earlier opt-out model. That proposal would have allowed AI developers to train models on copyright works by default unless rights were reserved. The Government has now confirmed that it no longer has a preferred option and that reform will take time to get right.

For businesses developing or deploying AI, the message is clear: using copyright-protected works for training engages existing UK copyright law and will generally require permission unless a narrow existing exception applies. Licensing remains the safest and most future-proof route.

AI training data licensing

Rather than weakening copyright, both reports point towards licensing as the mechanism through which AI development and creative rights can coexist.

The House of Commons report recognises that a licensing market for AI training is already emerging, particularly in sectors such as news, publishing, music and image libraries. It suggests that the Government should focus on supporting the conditions in which licensing can scale, rather than mandating new statutory schemes or exceptions.

This approach is echoed in the Government’s statement, which emphasises fair remuneration for creators and the need to ensure that independent and smaller creative organisations aren’t excluded from licensing opportunities.

For AI developers and businesses integrating AI into products or services, this reinforces the importance of thinking strategically about licensing models, supply chain risks and long-term commercial sustainability, rather than treating copyright as a short-term compliance issue.

AI transparency obligations & trust in the UK market

Transparency emerges as one of the strongest common themes across both Parliamentary reports and the Government’s response.

Rights holders consistently told Parliament that without meaningful transparency around training data, it’s extremely difficult to know whether works have been used to enforce rights or negotiate licences. The House of Commons report therefore places significant weight on input transparency and the development of standards, technical solutions and best practice.

The Government has confirmed that it’ll review the mechanisms available for creators to control how their works are used online. This review will consider standards, technical tools and approaches to input transparency and inform whether further intervention is needed.

For businesses, this points towards an environment in which record-keeping, auditability and data provenance will become increasingly important, particularly for larger developers and deployers of generative AI.

Digital replicas law

Both the House of Commons report and the ministerial statement highlight digital replicas as a growing area of concern. While AI-generated replicas can have legitimate and valuable uses, Parliament recognises the potential harms arising from the unauthorised use of a person’s voice, likeness or identity.

The Government has confirmed that it’ll consult on digital replicas, including the potential harms they cause and whether new protections are required. This sits alongside planned work on labelling AI generated content, with a particular focus on deepfakes and disinformation. Proposals are expected later in 2026.

For sectors such as music, film, television, games and digital media, these developments are likely to have direct relevance to product design, talent agreements and risk management.

Copyright & wholly AI-generated works

The House of Commons report also addresses the status of wholly AI-generated works. It confirms the Government’s view that copyright should incentivise and protect human creativity and that works generated entirely by AI without human authorship shouldn’t attract copyright protection.

This provides greater clarity for businesses using generative tools as part of a creative process, while preserving protection for AI assisted works where human creative input remains central.

Is all this enough to satisfy creators & AI companies?

The combined message from Parliament and the Government is one of caution and compromise.

For creators, the abandonment of the opt-out model and the emphasis on licensing, transparency and protections against harmful digital replicas will be welcomed. For AI companies, the lack of immediate legislative certainty and continued reliance on existing copyright law may present operational and commercial challenges.

Rather than a single sweeping reform, the UK is pursuing a phased approach. This involves further consultation, support for market-led licensing and targeted intervention where genuine gaps are identified. Further policy development and potential legislation are expected later in 2026.

Five practical steps for businesses using AI in the UK

In light of these developments, businesses using or developing AI should consider:

1. Reviewing training, fine tuning and retrieval datasets to identify copyright protected material.
2. Adopting a licensing-first approach and documenting permissions, scope of use and duration.
3. Strengthening record-keeping and audit trails around training data and data sources.
4. Reviewing contracts with suppliers and customers to address AI-related warranties, indemnities and transparency obligations.
5. Monitoring developments on digital replicas, labelling and transparency standards.

Failing to engage with these issues may expose organisations to legal risk, regulatory scrutiny and reputational harm — particularly where AI systems are trained or deployed at scale.

Impact on AI strategy & future regulation

Taken together, the House of Lords and House of Commons reports point towards a more cautious and considered approach to AI and copyright in the UK. Rather than sweeping legislative change, the direction of travel is towards licensing, transparency and targeted intervention.

This approach contrasts with developments in other jurisdictions such as the EU’s AI Act, which takes a more prescriptive regulatory stance. UK businesses operating internationally will need to account for these differences when designing AI systems and compliance frameworks.

For UK businesses, the opportunity lies in getting ahead of this evolving landscape. Embedding copyright compliance, transparency and fair remuneration into AI strategies now will reduce risk, support trusted partnerships and allow innovation to proceed with greater confidence.

Talk to us

Our intellectual property team advises businesses across the creative and technology sectors on the legal and commercial issues arising from the development, deployment and use of artificial intelligence.

We support clients across the full range of intellectual property rights (including copyright) and advise on the exploitation and licensing of AI‑enabled products and services. This includes structuring AI training data and content licensing strategies, as well as managing copyright infringement and enforcement risk where protected works are used within AI systems without permission.

Our team also advises on wider AI governance, transparency obligations and contractual protections, helping businesses to manage supply‑chain risks, regulatory scrutiny and reputational issues as the UK’s approach to AI and copyright continues to evolve.

By combining deep intellectual property expertise with a strong understanding of AI technologies, we help organisations to protect IP rights while enabling compliant, commercially sustainable innovation in the AI sector.

If you need support, give us a call on 0333 004 4488, send us an email at hello@brabners.com or complete our contact form.