Generative AI, EU Press Publishers’ Rights and the Australian News Bargaining Approach: Copyright and Competition Law as Enablers of Media Plurality[*]This article has been peer-reviewed.
By Kalpana Tyagi[**]Assistant Professor (Intellectual Property & Competition Law) and Founding and Managing Coordinator, The Innovator’s Legal Aid Clinic (TILC), Faculty of Law, Maastricht University. Email: k.tyagi@maastrichtuniversity.nl. I would like to thank the Intellectual Property Research Institute (IPRI) of Australia, Ben Hopper and Jake Goldenfein and all the participants at the IPRI roundtable for their valuable inputs. The proceedings of the roundtable are available here: https://www.youtube.com/watch?v=yB5yOneiOiE&t=680s. I would also like to extend her gratitude to Ariel Katz, Martin Kretschmer, Giovanni Ramello and all the other participants at the 2025 Annual Congress of the Society for Economic Research on Copyright Issues (SERCI) conference at the University of Hong Kong for their constructive inputs, and Grace Chan for finely managing the logistics. Errors, if any, are mine. This work covers the legal and technical developments until 1 June 2025.**
1. Introduction
The rapid rise of large digital platforms, such as Google and Meta, and their unauthorized use of news snippets and crawling adversely influence the incentives and remuneration received by press publishers, and journalists. To remedy this adverse outcome, in 2019, the EU’s Copyright in the Digital Single Market Directive (2019 CDSM) implemented in Article 15, a new neighbouring right, namely the press publishers right. The said right seeks to facilitate the sustainability of press publishing, and thereby, contribute to the constitutional ideal of an independent and pluralist press. As Google attempted to circumvent the right, the French Competition Authority (FCA), Autorité de la Concurrence intervened early on and imposed a series of measures against non-compliance by Google. The digital conglomerate however, still failed to comply with the said commitments and continued to use its strong bargaining position vis-à-vis press publishers. In March 2024, the FCA once again reprimanded Google, and fined €250 million for non-compliance with its 2022 commitments. Notably, Google had failed to inform the press publishers, let alone offer them the freedom to opt-out while training its foundational large language model (LLM), Bard (since, ‘Gemini’). Training LLMs is subject to text and data mining (TDM), an issue that falls under copyright and database rights. As generative AI becomes mainstream in journalism, a careful discourse on media plurality is more vital than ever, as an independent press is an enabler of a vibrant, and freely functioning democratic society. How does the EU copyright and related rights-driven solution, that is the neighbouring press publishers’ rights, and its enforcement by a national competition authority, the FCA, facilitate this constitutional ideal? In Australia, the issue is dealt with by a competition law-based instrument, namely the Australian News Media and Digital Platforms Mandatory Bargaining Code (NMBC).[1]In 2022, the Government of Canada introduced Bill C-188, the Online News Act. The said Act is similar to the Australian model. This research contribution though, concentrates only on the Australian approach, as my research stay and roundtable at Melbourne offered me an opportunity to better understand the Australian solution to the challenge faced by press publishers. For a rich, through and critical discussion on the Canadian approach, see Ariel Katz (2023) ‘Sedating Democracy’s Watchdogs: Critical Reflections on Canada’s Proposed Online News Act’ The Columbia Journal of Law & The Arts Vol. 46 No. 3 available here https://doi.org/10.52214/jla.v46i3.1123. The issue of the sustainability of press in an era of digital news consumption, thus, has been addressed through two distinct approaches – namely copyright and competition – across two different jurisdictions, the EU (with a civil law tradition) and Australia (with a common law tradition) respectively. In the US, likewise, there seems to be an understanding that the issue sits at the interface of copyright and competition, as proposals like, the Journalism Competition and Protection Act (JCPA) and the currently pending lawsuit, Helena World Chronicle[2]Helena World Chronicle LLC v Google LLC & Alphabet Inc. (filed 12/12/2023) Class Action Complaint: Jury Trial Demanded Case 1:23-cv-03677. On 11 December 2023, Helena World Chronicle filed a complaint against Google at the US District Court for the District of Columbia. As of 24th February 2026, the matter is still pending before the court. indicate. Does this bring together the traditionally distinct fields of law, intellectual property (copyright and related rights) and competition law closer together, as they join hands to strengthen the democratic fabric of media plurality in the age of generative AI? However, what exactly is meant by media plurality? Why does media plurality concerns remain central to generative AI? What changes as we transition our focus from the digital platforms to generative AI?
There are two ways to counter an unpleasant imprint – either defamation or alternatively, a more ‘sustainable argument with an alternative that remains convincing in the long run’.[3]This is a line of reasoning adopted by leading international law scholars, such as Prof. Eleanor M. Fox and Prof. Josef Drexl. Cf Eleanor M. Fox (1987) ‘The Battle for the Soul of Antitrust’ California Law Review 75 p. 917; Eleanor M. Fox (1981) ‘Modernization of Antitrust: A New Equilibrium’ 66 Cornell Law Review 1140, 1179 and Josef Drexl (2016) ‘Economic Efficiency versus Democracy: On the potential role of Competition Policy in Regulating Digital Markets in times of post-truth politics’ Max Planck Institute for Innovation and Competition Research Paper No. 16-16. In this research, I tread the latter path to look for alternatives that remain sustainable from a long-term perspective. This also espouses the need for cross-disciplinary thinking. With an inter-disciplinary insight from the world of journalism and the more recent technological developments (earlier, AI and now, generative AI and large language models, LLMs), does one fathom the stagnancies of traditional journalism. This also helps appreciate the evolution of consumer decision (more precisely, the reader’s decision) to replace the well-established and conventional sources of news with new and innovative digital platforms. This research contribution, not only presents the challenges in the news value chain, it also deliberates on remedies that can potentially resolve these challenges.
