Comments in Response to Consultation on Review of the 2019 Copyright in the Digital Single Market Directive

June 2026

Authors: Ahmed Nagy, Neil Yorke-Smith

Reviewers: Francisco Medeiros, Michel Beaudouin-Lafon

Editor: Tom Romanoff

Introduction

The European Commission’s call for evidence on the review of the 2019 Copyright in the Digital Single Market Directive is welcome. The related ‘targeted initiative’ aspires to foster a better copyright environment for European creativity and innovation. The initiative addresses an important and timely policy challenge: how to ensure that Europe’s copyright framework continues to support creators, researchers, public-interest institutions, users, and innovators in a rapidly changing technological environment.

This response is offered from a technology policy perspective, grounded in the public-interest orientation of the computing profession. The objective is not to privilege one sector over another, but to support a balanced, evidence-based, technically realistic, and rights-respecting copyright framework. Such a framework should protect the legitimate interests of creators and rights holders, while enabling scientific research, responsible AI development, cultural access, education, and European competitiveness.

Overall assessment

The Commission is right to examine the practical effects of the 2019 Copyright in the Digital Single Market Directive and to consider targeted adjustments where recent technological developments have created new uncertainties. Generative AI, large-scale data processing, online platforms, live-content piracy, and cross-border research collaboration have changed the practical conditions under which copyright is exercised and enforced.

The review should remain proportionate and modular. AI training, AI-generated outputs, impersonation of performers, live-event piracy, scientific publishing, and remuneration for recorded music are all related but distinct policy problems. They should not be addressed with a single broad, undifferentiated legal response. The Commission should distinguish clearly between the different technical and legal stages at which copyright-relevant acts may occur.

We recommend that the framework should be informed and guided by five principles:

1. Rights balance: copyright protection should be balanced with freedom of expression, access to knowledge, scientific research, privacy, cultural participation, and innovation.
2. Interoperability and legal certainty: licensing, opt-out, transparency, and remuneration mechanisms should be machine-readable, cross-border and easy to implement.
3. Technical accuracy: legal obligations should reflect how AI systems, datasets, platforms, and digital distribution systems operate, while being open to future technical developments.
4. Proportionality: obligations should be scaled according to actor size, risk, capability, and market impact, with special attention to SMEs, start-ups, researchers, and public-interest institutions.
5. Accountability with safeguards: enforcement tools should include transparency, due process, appeal rights, and protection against over-blocking or excessive surveillance.

Copyright and generative AI

Generative AI raises legitimate concerns for creators and rights holders. These include a lack of visibility in training data, uncertainty about the effects of rights reservations, difficulty licensing works at scale, potential market substitution, and AI-generated imitations of personal characteristics or performances.

At the same time, AI development also depends on access to data, research freedom, technical experimentation, and legal certainty. Europe should therefore avoid a framework that protects creators only by making responsible AI development impractical, or that supports AI innovation only by weakening creative rights.

The Commission should distinguish between the five stages listed below:

• Data collection. Defines how content is accessed, scraped, purchased, licensed, or otherwise obtained.
• Dataset curation. Concerns include filtering, de-duplication, exclusion of illegal sources, respect for opt-outs, quality assurance, and documentation.
• Model training. Raises questions about whether the relevant uses of content fall within text and data mining exceptions, licensing arrangements, or another lawful basis.
• Model deployment. Defines transparency, user controls, risk estimation, safety, monitoring, and compliance mechanisms.
• Output generation: Concerns potentially infringing outputs, imitation of protected expression, deceptive impersonation, possible market substitution, and appropriate remedies.

This distinction is important because no single legal mechanism can govern all stages fairly. Training-data transparency, licensing, remuneration, output liability, and performer protection should each be designed for the specific technical stage they address.

Transparency and data governance

Transparency should be strengthened, but in a tiered and proportionate way. Public summaries of training content are useful but may be insufficient if they are too generic. At the same time, requiring full disclosure of datasets could create risks for trade secrets, cybersecurity, privacy, and copyright itself.

A balanced framework should include three levels of transparency:

Public transparency: AI providers should publish meaningful summaries of training-content categories, data sources, exclusion practices, opt-out compliance measures, and high-level dataset governance processes.

Regulatory transparency: competent authorities should be able to request more detailed documentation where necessary for enforcement, while protecting trade secrets, personal data, and security-sensitive information.

Controlled-access transparency: in disputes or audits, qualified independent bodies should be able to examine more detailed information under confidentiality safeguards.

This approach would support creator rights without forcing disclosure of full datasets to the public.

Licensing and remuneration

Licensing mechanisms that are practical, interoperable, and accessible should be supported. Licensing should not become a procedure that only large platforms, large collecting societies, or large rights aggregators can navigate.

The revised Directive should encourage:

• Machine-readable rights reservations and licensing signals;
• Standardised metadata for rights ownership and permitted uses;
• Interoperable registries where appropriate;
• Dispute-resolution mechanisms for licensing conflicts;
• Low-cost pathways for SMEs, researchers, and public-interest users;
• Safeguards against excessive transaction costs;
• Mechanisms for individual creators and small rights holders to participate meaningfully.

Where remuneration mechanisms are considered, they should be evidence-based and should avoid double payment, market concentration, or regulatory designs that disadvantage European start-ups.

Performer protection and impersonation

The revised framework should consider targeted protection against commercially harmful or intentionally deceptive AI-generated impersonation, particularly where a performer’s voice, image, likeness, or distinctive performance is reproduced or simulated without consent and is likely to mislead audiences or cause material harm. Such protection should preserve lawful uses, including parody, criticism, quotation, education, research, accessibility, and other legitimate transformative uses.

Research, open science, and secondary publication rights

The revised Directive should give high priority to harmonising and strengthening copyright rules for research. Fragmented national implementation creates uncertainty for cross-border research and undermines the European Research Area.

The Directive should protect text and data mining, computational analysis, reproducibility, preservation, and secure sharing of research materials for lawful scientific purposes. These rules should apply clearly across borders and should support public-interest research, including AI research, digital humanities, cybersecurity research, and data-intensive science.

The Directive should consider establishing a harmonised EU-level secondary publication right for publicly funded research. While some Member States, funders, and institutions already provide open-access mechanisms or rights-retention requirements, these arrangements remain fragmented and do not provide all researchers with a clear and consistent legal basis to make accepted manuscripts available through open repositories.

Researchers should have a clear, non-waivable right to make the accepted manuscript of publicly funded research available in an open repository after a defined and proportionate embargo period, subject to carefully specified exceptions for legitimate confidentiality, security, privacy, and intellectual-property considerations.

Implications for SMEs and European competitiveness

The review should avoid increasing compliance burdens in ways that only the largest actors can absorb. If copyright compliance becomes too complex or expensive, Europe may unintentionally strengthen dominant non-European platforms and weaken European start-ups, public-sector innovators, research organisations, and SMEs.

The revised Directive should therefore consider proportional obligations, regulatory sandboxes, standard templates, technical guidance, open standards, and shared compliance infrastructure. In particular, open and interoperable tools for rights metadata, licensing, opt-out management, dataset documentation, and audit trails should be supported.

Recommendations

The revised Directive should:

1. Adopt a modular approach that separates AI training, AI deployment, AI outputs, performer impersonation, piracy, music remuneration, and research access;
2. Implement tiered transparency for AI training content, combining public summaries, regulatory access, and controlled independent audit mechanisms;
3. Develop interoperable and machine-readable mechanisms for rights reservations, licensing, and metadata;
4. Ensure that licensing and remuneration mechanisms are accessible to individual creators, SMEs, researchers, public-interest institutions, and European start-ups;
5. Strengthen protection against harmful AI impersonation while preserving legitimate expression, parody, criticism, education, accessibility, and research;
6. Improve enforcement against professional piracy of live content, but only with due process, proportionality, transparency, and redress safeguards;
7. Harmonise research exceptions across the EU and support a secondary publication right for publicly funded research;
8. Ensure that any new copyright obligations for general-purpose AI providers complement, rather than duplicate, the transparency and copyright-compliance duties established under the EU AI Act; protect personal data and confidentiality in accordance with GDPR; and support the European Research Area by providing clear, cross-border rules for lawful text and data mining, secure research access, and secondary publication of publicly funded research. Mechanisms should be machine-readable, interoperable across borders, and easy to implement.
9. Create a structured evidence base before legislative intervention, including economic impacts on creators, SMEs, AI providers, researchers, cultural institutions, and consumers and users.

Conclusion

Europe needs a copyright framework that protects creativity and supports innovation. These goals are not contradictory. A well-designed framework can strengthen creators’ rights, improve licensing markets, support trustworthy AI, protect performers, combat professional piracy, and expand access to scientific knowledge.

Technically informed and rights-balanced reforms should be pursued. The guiding objective should be a European copyright environment that is fair to creators, workable for innovators, protective of users and researchers, and aligned with the public interest.

References Consulted

European Commission. “Call for Evidence: Report on the Review of the Copyright in the Digital Single Market Directive / Targeted Initiative for a Better Copyright Environment for European Creativity and Innovation.” Have Your Say, 13 May 2026, https://ec.europa.eu/info/law/better-regulation/have-your-say/initiatives/18173-Targeted-initiative-for-a-better-copyright-environment-for-European-creativity-and-innovation-_en. Accessed 23 June 2026.

European Commission. “Guidance on Article 17 of Directive (EU) 2019/790 on Copyright in the Digital Single Market.” EUR-Lex, Communication COM(2021) 288 final, 4 June 2021, https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52021DC0288. Accessed 23 June 2026.

European Commission, Directorate-General for Research and Innovation. “Economic Impact of Options for Improving Access to and Reuse of Research Results, Publications and Data for Scientific Purposes.” Research and Innovation, 13 May 2026, https://research-and-innovation.ec.europa.eu/events/upcoming-events/economic-impact-options-improving-access-and-reuse-research-results-publications-and-data-scientific-2026-05-13_en. Accessed 23 June 2026.

European Commission, Directorate-General for Research and Innovation. “Boosting Knowledge Circulation: New Study on Economic Impact of Possible Measures.” Research and Innovation, 30 Apr. 2026, https://research-and-innovation.ec.europa.eu/news/all-research-and-innovation-news/boosting-knowledge-circulation-new-study-economic-impact-possible-measures-2026-04-30_en. Accessed 23 June 2026.

European Parliament and Council of the European Union. Directive (EU) 2019/790 of 17 April 2019 on Copyright and Related Rights in the Digital Single Market and Amending Directives 96/9/EC and 2001/29/EC. Official Journal of the European Union, L 130, 17 May 2019, pp. 92–125. EUR-Lex, https://eur-lex.europa.eu/eli/dir/2019/790/oj/eng. Accessed 23 June 2026.

European Parliament and Council of the European Union. Regulation (EU) 2016/679 of 27 April 2016 on the Protection of Natural Persons with Regard to the Processing of Personal Data and on the Free Movement of Such Data, and Repealing Directive 95/46/EC (General Data Protection Regulation). Official Journal of the European Union, L 119, 4 May 2016, pp. 1–88. EUR-Lex, https://eur-lex.europa.eu/eli/reg/2016/679/oj/eng. Accessed 23 June 2026.

European Parliament and Council of the European Union. Regulation (EU) 2024/1689 of 13 June 2024 Laying Down Harmonised Rules on Artificial Intelligence and Amending Regulations (Artificial Intelligence Act). Official Journal of the European Union, 12 July 2024. EUR-Lex, https://eur-lex.europa.eu/eli/reg/2024/1689/oj/eng. Accessed 23 June 2026.

PDF available here.