Dr Trapova speaks to Reuters Institute about AI, copyright and journalism

Dr Alina Trapova, Lecturer in Intellectual Property Law at UCL Laws and Co-Director of the UCL Institute of Brand and Innovation Law (IBIL), was interviewd by Gretel Kahn for an article published by the Reuters Institute for the Study of Journalism.

At a time when some media organisations are actively challenging AI companies in court, the article examines whether existing copyright laws still offer adequate protection for news publishers. Dr. Trapova outlines some of the complexities surrounding use of copyright-protected material as AI training data, and she also explains why, while concerns over wholesale infringement are widespread, legal clarity is still lacking. There is still no definitive court ruling determining whether the act of training large language models (LLM) complained of amounts to copyright infringement.

“The general feeling in all of these discussions is that there is a violation of copyright. But it has to really go down again to how each individual [AI] system works, what kind of copies are retained, for how long, what is actually there in these AI models and in the training data they use.” 

Dr Trapova highlights that because copyright law automatically protects creative works upon creation, without requiring registration, it lacks the certainty provided by registered intellectual property rights, including patents and trade marks. This uncertainty often pushes stakeholders towards licensing agreements rather than litigation. 

“[Journalism] is a creative industry which thrives on reputation, and it has a very complex business model as well ... . So regardless of whether copyright is clear or not, and in many cases it is really not a clear legal framework, people prefer to have a document signed and the license established to say ‘Yes, you have allowed me to train or you have not allowed me to train for this and that purpose.’ Certainty comes with that agreement, because the laws are not clear.” 

The European Union has attempted to address some of these concerns with the 2024 AI Act, the world’s first piece of legislation regulating AI. While the Act covers a broad range of topics, it does include copyright-related provisions. According to Dr. Trapova, this means that any AI system deployed or used in the EU must comply with EU copyright laws, regardless of where it was trained or where the company is based. 

“I think it’s clever. I’m not really sure whether it plays out very well in practice. It is drafted in a complicated way, and it’s one little provision,” says Dr. Trapova. “But the EU has learned from its past mistakes in digital lawmaking and it has produced a code of practice that is related to the AI Act. There’s a first draft of it, which was written by independent experts. It’s open now for comments so people can actually amend it, and that code of practice will give some details to how this aspect will play out.” 

As litigation between news organisations and AI companies continue, the question remains: will existing copyright laws be enough, or will new legislation be required? Dr. Trapova suggests that in the UK, legislative developments may be on the horizon, as copyright holders increasingly voice their concerns. While it takes time for a proposal to become law, she notes that once policymakers signal a clear direction, stakeholders may gain confidence in the legal framework. 

You can read the full article here.