Intellectual Property: Artificial Intelligence

Samantha Niblett Excerpts

Wednesday 23rd April 2025 Westminster Hall

[Samantha Niblett (South Derbyshire) (Lab)]

It is an honour to serve under your chairship, Ms McVey, and I thank my hon. Friend the Member for Bury North (Mr Frith) for securing this crucial debate—a great way to spend his birthday.

I want to draw attention to the fact that my partner works in publishing. My background is not in publishing or the creative industries, but in the data and tech industry, and I am excited by what it can help us to achieve. I rise to speak not only about artificial intelligence, but about the kind of country we want to be in the decades ahead. We are at a crossroads. The United Kingdom is positioning itself as a global leader in AI, and rightly so—AI has the potential to transform every sector of our economy, from healthcare to transport and from education to climate science—but, as we drive innovation forward, we must also ensure that our values are aligned.

If we want to be the best place in the world to start, scale and grow an AI company, we must also be the most trusted. Trust is not built through ambition alone; it is built through transparency, fairness, respect for the rule of law and the protection of intellectual property. At present, many of the most powerful AI models have been trained on vast quantities of copyrighted material—books, articles, art, music and software—and often without the consent, knowledge or compensation of the people who created it. That is not hypothetical; it is happening now. In the last week, we have seen some big names in tech say that they want IP law to be ripped up altogether. I wonder how they would feel if tech firms with deeper pockets and bigger legal teams than theirs decided to simply take what they have. Pulling up the ladder of IP law may no longer seem such an attractive prospect.

Let us be absolutely clear: copyright is not an inconvenience or a technicality. It is the legal and moral framework that ensures that creators are rewarded for their contributions. Some people will argue that requiring AI companies to comply with copyright law would hinder innovation, but the opposite is true: legal certainty enables innovation. What stifles innovation is a regulatory vacuum where only the largest, best resourced firms can afford to operate in the grey areas of the law. We should not accept a model where creativity becomes collateral damage in the pursuit of speed or short-term profit for shareholders. Governments must prioritise the concerns of citizens.

Let us establish a clear, transparent and enforceable framework for the use of copyrighted content in AI training, and let us not waste any time or kick this can down the road by waiting for the results of the copyright and AI consultation. Let us act now. Let us amend the Data (Use and Access) Bill to ensure that companies providing generative AI services in the UK must comply with UK copyright law and be transparent about the data fuelling their models. After all, they should know what data they are utilising. It can be done; companies including Adobe, 273 Ventures and Flawless AI are doing it. Experts have made it clear that no opt-out models work. We have an opportunity to be world leaders—let us lead.