Skip to content 
On 6 March 2026, the House of Lords Communications and Digital Committee published what may be the most consequential legal report on intellectual property in a generation. Titled AI, Copyright and the Creative Industries, the report draws a clear line in the sand: the UK’s creative sector must not be sacrificed on the altar of speculative AI gains. This report holds significant importance for IP owners, rights holders, creative professionals, and businesses that rely on original content, and the upcoming weeks will determine the government’s response.
At Lawdit Solicitors, we have been advising clients on IP protection for over two decades. We have watched the AI debate develop with growing concern. The Lords committee has now said, loudly and clearly, what many of our clients have been telling us for months: the current situation is untenable, and the law needs to catch up.
Here is what the report says, what it means in practice, and—most
importantly—what you should be doing right now to protect your intellectual property.
What Did the House of Lords Actually Say?
The Committee’s central finding is stark. Generative AI systems—the tools that produce text, images, music, and video on demand—are trained on vast quantities of human-created content. Much of that content is copyrighted. Much of it was used without the creators’ consent or payment. The Committee describes its findings as a “clear and present danger” to the UK’s creative industries.
The report is not, crucially, an attack on AI itself. The Committee acknowledges the technology’s potential and accepts that it will play a significant role in the UK’s economic future. What it objects to is the mechanism by which AI companies have been acquiring the training data they need: by hoovering up copyrighted material from across the internet without licensing it, without disclosing it, and without remunerating the people who created it.
The Committee frames the government’s choice as a binary one. The UK can either become a world-leading home for responsible, licensing-based AI development — or it can drift toward a situation where a small number of US-based technology firms extract enormous value from British creative work while the creators themselves see nothing in return. The report calls the latter path “a race to the bottom that does not serve UK interests.”
The Text and Data Mining Problem
Central to the debate is the question of text and data mining (TDM) exceptions. TDM is the process by which AI models are trained: they consume and analyse vast datasets, extracting patterns and relationships that allow them to generate new content. Under current UK copyright law, there is a limited TDM exception for non-commercial research. The controversy has been whether to extend this restriction to commercial AI training.
The government consulted on four options, ranging from leaving the law unchanged to creating a broad exception that would allow commercial AI training with no opt-out for rights holders. Its initially preferred option — a commercial TDM exception with an opt-out mechanism — attracted fierce opposition from across the creative sector. Artists, musicians, authors, publishers, filmmakers, and photographers argued that an opt-out model places the burden squarely on rights holders: unless you actively register your objection, your work can be used freely by AI companies to generate commercial products that compete directly with you.
The Lords Committee agrees. It recommends the government formally rule out any new commercial TDM exception with an opt-out mechanism. Instead, it advocates for a licensing-first approach: AI companies should negotiate and pay for access to the content they use, just as broadcasters license music or publishers license images.
What the Report Recommends—In Plain English
The Committee makes several specific recommendations that IP owners should understand:
1. No opt-out TDM exception. The government should publicly and definitively rule out any mechanism that would allow commercial AI companies to use copyrighted material unless rights holders actively opt out. The burden of protection should not fall on creators.
2. Mandatory transparency. AI developers should be legally required to disclose what training data they have used, to a standard that is sufficiently detailed and granular for rights holders to determine whether their work has been ingested. The current opacity—in which AI companies decline to reveal their training datasets—must end.
3. Protection against digital replicas and style imitation. This is significant. The Committee calls on the government to introduce legal protections against unauthorised digital replicas—deepfakes, essentially—and against AI outputs that imitate a creator’s style without permission. UK law currently has significant gaps here. A musician cannot, in most circumstances, prevent an AI from generating music “in the style of” their catalogue. The Committee wants that to change.
4. A licensing market that actually works. Rather than weakening copyright, the government should focus on building the infrastructure for a functioning licensing market. This means technical standards for data provenance, watermarking and content authentication, and frameworks that allow rights holders and AI developers to transact fairly.
5. Sovereign AI development. The Committee urges investment in UK-developed AI models trained on licensed content — models that could offer a transparent, rights-respecting alternative to the opaque, US-based systems that currently dominate the market.
Why This Matters for Your Business
The economic stakes are not trivial. The UK creative industries contributed £124 billion to the economy in 2023 and employ 2.4 million people. By contrast, the AI sector employed just 86,000 people and contributed £12 billion in 2024. The Committee’s implicit point is clear: the creative sector is not a marginal interest group — it is one of Britain’s most important industries, and it deserves protection proportionate to its economic weight.
But beyond the macroeconomic picture, this debate has direct implications for individual IP owners across a range of sectors.
If you work as a photographer, illustrator, or visual artist, it’s possible that AI image generators have already used your images for training. Without transparency requirements, you have no way of knowing. The report’s call for mandatory disclosure is directly relevant to you — and the window to register your concerns with the government is right now.
If you are a musician, composer, or music publisher, the style imitation gap in UK law is a real and present risk. AI tools can already generate music that sounds convincingly like a named artist. The absence of a personality right or style protection in UK law leaves you exposed. The Committee’s recommendation to close this gap is welcome — but it is a recommendation, not yet law.
If you are a software developer or technology business, you need to understand how the licensing landscape may change. If a commercial TDM exception is ruled out and a licensing-first regime takes hold, the cost and legal complexity of training AI models on third-party content will increase. We may need to revisit business models that assumed free access to internet-scraped training data.
If you are a content creator, publisher, or media business, the Committee’s emphasis on transparency and provenance standards is important. Technical standards like the Coalition for Content Provenance and Authenticity (C2PA) specification, which includes watermarking, fingerprinting, and machine-readable metadata, could help you see if your content has been used and protect your rights. Comprehending these standards now positions you at the forefront of the industry.
What Happens Next
The government was required by the Data (Use and Access) Act to publish an economic impact assessment and a progress update on its copyright consultation by 18 March 2026. Following the Lords report, pressure on ministers has intensified considerably. Secretaries of State Liz Kendall and Lisa Nandy have already indicated that the government’s earlier preference for a TDM opt-out mechanism was a mistake — describing the current moment as a “reset”. The government is also planning to trial a Creative Content Exchange, testing commercial licensing models with institutions including the National Archives, Historic England, and several major museums, with a view to launching an operational platform by summer 2026.
The political direction of travel, in other words, is towards greater protection for rights holders — but the detail of how that protection is implemented remains to be determined. The next few months will be critical.
What You Should Do Now
This is not a moment for passivity. If you own intellectual property — whether registered trademarks, copyright in creative works, design rights, or software — there are practical steps you should take immediately.
Audit your IP portfolio. Do you have a clear record of what you own, when it was created, and where it has been published? If your work is publicly accessible online, it is potentially at risk. A thorough IP audit is the foundation of any protection strategy.
Review your licensing arrangements. If you license your content to third parties — platforms, publishers, aggregators — check the terms carefully. Do those licences permit sublicensing for AI training purposes? Many standard content licences predate the current AI debate and may not address the issue at all.
Document your creative process. In the event of a dispute — whether with an AI company or a competitor using AI-generated imitations of your work — evidence of the creative process matters. Maintain records of drafts, correspondence, and the development of your work.
Take advice on trademark and personality rights. For individuals and brands whose identity, voice, or likeness could be replicated by AI, the current absence of robust personality rights in UK law is a genuine vulnerability. Registered trademarks can offer some protection — but understanding the limits of that protection, and what additional steps are available, requires specialist advice.
Stay engaged with the consultation process. The government’s position on AI and copyright is not yet finalised. Rights holder organisations, industry bodies, and individual businesses can still make their voices heard.
Lawdit’s View
We believe the House of Lords Committee has got this right. The UK’s copyright framework is not broken — it has served creators, businesses, and the broader economy well for decades. What is needed is not a weakening of that framework to accommodate a handful of large technology companies, but an extension and modernisation of it to ensure that the rights it protects remain meaningful in an age of generative AI.
The licensing-first approach the Committee advocates is practical, fair, and consistent with how intellectual property has always worked: if you want to use someone’s creative work commercially, you pay for it. There is no principled reason why AI training should be an exception to that rule.
The coming weeks will tell us whether the government agrees. Whatever happens, the legal landscape around AI and copyright is changing fast—and IP owners who understand what is at stake and take steps to protect their position now will be far better placed than those who wait.
If you have questions about how the AI copyright debate affects your intellectual property, or you would like advice on protecting your creative works, trademarks, or business content, please contact Lawdit Solicitors. Our specialist IP team has been advising clients on copyright, trademarks, and commercial disputes since 2001.
Book a Consultation | Intellectual Property Services | lawdit.co.uk
