**Summary Abstract: UK Experts Warn Against AI 'Opt-Out' Model for Creative Content** A new report from the University of Cambridge, a collaboration between the Minderoo Centre for Technology and Democracy, the Bennett Institute for Public Policy, and ai@cam, cautions the UK government against adopting an 'opt-out' model for AI data mining of copyrighted creative works. The report argues that this approach could stifle the growth of new talent in the creative sector and compromise the livelihoods of artists, musicians, and writers, particularly those just starting out. The UK's creative industries, which contribute approximately £124.6 billion or 5.7% to the national economy, could be significantly harmed by such a policy, as it would allow AI companies, often based offshore, to scrape and reuse artistic content without adequate compensation. **Key Events and Stakeholders:** - **Report Publication:** The University of Cambridge has published a report on the implications of AI data mining in the creative sector. - **Government Proposal:** The UK government is considering an 'opt-out' model for AI companies to scrape copyrighted works unless creators explicitly opt out. - **Collaborating Institutions:** The Minderoo Centre for Technology and Democracy, the Bennett Institute for Public Policy, and ai@cam. - **Lead Authors and Experts:** Prof Gina Neff, Prof Diane Coyle, and Dr Ann Kristin Glenster. **Current Landscape:** - **Economic Contribution:** The creative industries are a vital part of the UK economy, with the video games sector alone contributing £5.12 billion in 2019. - **AI Adoption:** While AI has the potential to foster innovation and new creative companies, its adoption in these industries is still in its early stages, and there are significant uncertainties about its impact. - **Legal Frameworks:** The current UK copyright law automatically protects creators, but some AI companies have attempted to exploit 'fair dealing' provisions for research and reporting purposes, which are often commercially driven. **Report Recommendations:** 1. **Maintain Automatic Copyright Protection:** The report strongly recommends maintaining the legal foundation that automatically safeguards copyright, rather than shifting to an opt-out model. 2. **Clarify Human Authorship:** It argues that only human authors should hold copyright, even when AI is heavily involved in the creative process. 3. **Commission Research:** The government should commission research to engage directly with creatives, understanding the benefits and harms of AI adoption, and use this to inform supportive policies. 4. **Implement the Beijing Treaty:** The researchers call for the full adoption of the Beijing Treaty on Audio Visual Performance, which the UK signed over a decade ago but has not yet implemented. This treaty would provide performers with economic rights over all reproductions, distributions, and rentals of their work. 5. **Transparency and Licensing:** They propose mandatory transparency requirements for AI training data and standardized licensing agreements that properly value creative works. This would help ensure that AI companies are held accountable and that creative professionals receive fair compensation. **Concerns:** - **Burden on New Talent:** The opt-out model places an unfair burden on up-and-coming creatives who may lack the skills and resources to navigate complex legal requirements. - **Unregulated Use:** Unregulated use of generative AI could lead to the exploitation of creative content by foreign companies, undermining the UK's creative sector. - **Uncertainty and Hesitancy:** The current lack of clarity about copyright infringement and the regulation of training data use is hindering the development and deployment of generative AI for public benefit in the UK. AI companies are hesitant to invest due to the uncertain legal landscape. - **AI-Generated Work:** The report also addresses the issue of copyright for AI-generated work, concluding that AI itself cannot hold copyright. However, guidelines are needed to ensure fair compensation for artists whose work and names are used in AI prompts. **Conclusion:** The Cambridge experts emphasize that the UK government should prioritize policies that enable the creative industries to benefit from AI while protecting the rights and interests of creators. They argue that the 'opt-out' model is not in the spirit of copyright law and could lead to a lose-lose situation, where creative professionals are not fairly compensated and AI companies are hesitant to invest. Instead, a clear and equitable framework that balances the interests of both sectors is essential for fostering growth and innovation in the UK.