As the UK explores how to harness artificial intelligence (AI) to fuel growth and global competitiveness, the discussion around copyright is at a critical juncture. At a recent panel hosted by the Center for Data Innovation, Benjamin White, researcher at Bournemouth University and Knowledge Rights 21 (KR21) co-founder and national coordinator for the UK, brought clarity to the growing concerns around overly restrictive copyright approaches that risk sidelining the huge contribution that R&D, and the research-intensive industries make to the UK economy.
This blog – part of a series on recent developments around AI – explores the developments in the UK.
What’s at Stake
Ben White reminded the audience that copyright law, though often seen through the lens of the entertainment sector, touches every sector of the economy. From healthcare to logistics, from finance to education, any sector using machine learning and data analysis is likely to be impacted by narrow and rigid interpretations of copyright law.
Yet current UK policy discussions seem disproportionately focused on the so-called “1%” of creative outputs, like music and film, ignoring the broader implications for sectors where AI is driving real-world productivity and research breakthroughs. “The rules that affect singers also affect scientists,” Ben said. He then clarified that introducing requirements that force the relicensing of materials that have already been licensed or are legally accessible, such as those proposed under Option 3 of the UK Consultation, would create substantial barriers to scientific progress and economic competitiveness.
Reinforcing a Public Interest Perspective
Ben drew on KR21’s mission and his own background in licensing to highlight a core issue: publicly funded knowledge is being paid for multiple times over. He set out the current financial burden to research as follows:
- Initial (public) funding: Taxpayers finance the majority of research from the outset, covering around 80% of its costs.
- Access costs: individuals and institutions must then pay again to access academic journals.
- Sharing costs: Additional fees apply when trying to share this knowledge internally within organisations like universities, the NHS, or private companies.
- Potential data mining costs: Finally, there could be an obligation to pay once more to mine it for facts and data, which aren’t even protected by copyright in the first place.
Such quadruple-dipping, as he framed it, is not just inefficient, it’s fundamentally unjust and anti-innovation. He referenced how Japan’s “non-enjoyment” AI exception enables training on copyrighted material as long as it’s not substantially similar to the inputs – a model that balances creator protection with real-world technological development.
A Better Path Forward
From his own research at Bournemouth’s Centre for Intellectual Property Policy & Management, Ben called for the UK to embrace a flexible, principle-based copyright regime — akin to those in the US, Japan, Singapore and South Korea — that recognises:
- Text and data mining (TDM) is a core research method, not a threat.
- A requirement to relicense materials that an organisation already has licensed or has legal access to cannot scale to meet AI’s needs, especially for publicly funded data.
- Innovation should not be throttled by outdated legal frameworks.
“This isn’t a zero-sum game,” he emphasised. “Countries like Japan, South Korea and the US with flexible copyright regimes still have thriving cultural sectors [with huge exports]. We can have both innovation and artistic vibrancy if we modernise our laws smartly.”
Why This Matters Now
With the UK’s AI strategy setting bold ambitions, policymakers must choose whether to enable innovation or entrench inefficiency. KR21 echoes Ben’s warning: overly cautious copyright policy risks outsourcing AI leadership to jurisdictions with better frameworks.
The UK doesn’t need more legal ambiguity or more barriers. It needs clarity, flexibility, and fairness, and that starts with recognising that the real threat to creativity is not AI, but the legal uncertainty that stifles it.
KR21 Recommendation
If the UK is serious about supporting AI in critical sectors like health, science, and education, it must extend its TDM exception to commercial uses and avoid the trap of opt-outs and licensing models that the EU itself has just woken up to. They are both impractical and economically harmful.
Read other recent publications about copyright, research and AI:
- More Than a Loophole: EUIPO Report on Generative AI and Copyright Out of Sync With the EU Research Agenda
- Focusing on the Future: Why AI Needs Freedom to Flourish in the UK
- A False Dichotomy, a Real Risk – Review of the AI Continent Proposals
- Boosting AI innovation in the UK
- Undermining Scientific Research
- Exploring the Intersection of AI, Text and Data Mining, and Copyright: A Reflection on the Stockholm Conference
12 May 2025
