Strengthening European Public Service Media in the Era of AI
Yannis Papadopoulos
27 July 2026 | Issue No. 194
Dr Yannis Papadopoulos is the President of the Board of Directors at the Hellenic Broadcasting Corporation – ERT. In June 2026 he was unanimously elected on the Executive Board of the European Broadcasting Union (EBU).
Abstract
This paper discusses current challenges facing European Public Service Media from a theoretical and practical perspective. While it acknowledges risks, it sees technology, in particular AI tools, as an asset which – if used properly – will be beneficial. The paper focuses on the successful example of the ‘European Perspective’, an initiative of the European Broadcasting Union (EBU) of which ERT is an active member and a partner, and considers deeper collaboration at EU level as a precondition to strengthen the role of Public Service Media.
Introduction
Public Service Media have traditionally functioned as pillars of democracy in Western societies. Deregulation policies in the 1980s and the 1990s, and later the advent of the internet, have not changed this mission but have certainly impacted on the nature of media culture. Commercialism, and the marketisation and digitalisation of the news has sparked an academic debate about the contribution of Public Service Media to democracy and has led scholars to focus on aspects such as their management structure, market impact, funding and quality of the news to evaluate their performance. The comparison of different systems, namely Public Service Media and market-driven media, has offered interesting insights concerning, among other things, the greater prominence the former devote to the news and their ability to contribute to smaller knowledge gaps between advantaged and disadvantaged people.1 In his 2017 account The Democratic Value of News Stephen Cushion stressed the significance of Public Service Media notwithstanding challenges.2
The academic discussion about Public Service Media, although not new, is currently being updated with the ongoing endeavour to interpret the cataclysmic penetration of Artificial Intelligence (AI) applications into the media industry. Generally speaking, AI is considered by scholars as both an opportunity and a threat. AI can help Public Service Media remain relevant but only by abiding by a set of rules that promote their values.3 Not only scholars but also media professionals are engaged with this necessary balance. Since 2024, the European Broadcasting Union (EBU) has expressed concerns about the gatekeeping role of AI tools that arguably restrict the access of European citizens to media services they find relevant.4 Hence, the Union supports its members to effectively implement AI tools in the best interest of their audience.
AI, the Media Reliability Crisis and Practical Solutions
To better outline the new landscape for Public Service Media in the era of AI a reference to the work of communication theorist Marshall McLuhan is useful. In 1964, McLuhan argued that ‘the medium is the message.’5 In his words the characteristics of the medium could determine the impact of the message on society. More than six decades later, McLuhan’s argument finds resonance but in a different context. The medium itself still matters but operates differently in the era of technology. The medium can now create the message by regularly relying on AI. This is a seismic shift, a structural transformation in the way information is generated, distributed, and understood. For the concept of journalism itself as well as for media professionals committed to serving the public interest, this shift challenges basic principles.
The penetration of AI applications in the generation of news goes together with an already existing crisis in the trustworthiness of journalism. This crisis does not display any signs of improvement this year. On the contrary, the Reuters Institute 2026 Digital Report shows that trust in news has dropped to 37%, the lowest level since 2015.6 The same study also demonstrates that the usage of AI chatbots for news consumption has increased to 10% of all audiences, while younger people are eager to use AI chatbots than older ones. Although this percentage is not particularly high, the trend has to be noted. It is therefore interesting that academic research, which has focused on analysing AI applications in the newsroom, has not extensively explored their usage by the audience.7
The objective of Public Service Media is to support the daily work of journalists as well as address the existing reliability crisis. AI tools can be a powerful ally when used wisely. What is required is transparency with regard to how and to what extent they are used. AI already plays a key role in supporting collaborative journalism across Europe. The ‘European Perspective’ constitutes a characteristic example.8
This is a pan-European newsroom established in 2021 by the EBU and co-financed by the EU. There are currently 22 participating organisations and partners in 18 countries including the Hellenic Broadcasting Corporation – ERT – (since 2024). Journalists feed their stories to a digital news hub, and editors select those they prefer and publish them immediately on their websites and apps. The European audience is then able to access stories (over 3,000 daily) which are the result of this innovative collaboration of European public service media. What makes the ‘European Perspective’ special, among other things, is that it uses AI tools to support the production, adaptation, and distribution of stories. Articles are translated into multiple languages using the EBU AI-assisted EuroVox tool, while the EBU automatic content recommendation tool, PEACH, helps surface stories where they are most relevant.
The ‘Fourth Estate’ mission of the media is not subject to reversal or change because of automation brought by AI tools. This commitment becomes more significant in a period during which disinformation risks are growing. To no one’s surprise the EU is building up capabilities to tackle Foreign Information Manipulation and Interference.9 Public Service Media function as a critical defence mechanism against these threats. The EU has responded by strengthening the European legal framework through initiatives such as the AI Act and the European Media Freedom Act. Public Service Broadcasters, including ERT. are in the process of complying with new rules. The EBU, for its part, calls for AI governance frameworks which recognise public service media as anchors of trust and demand accountability from AI companies involved in the generation of news.10
AI tools can be of crucial assistance not only in facilitating the journalistic work, but in helping with the identification of manipulated images, synthetic media, and deepfakes. They can detect patterns of misinformation, coordinated manipulation, and unsafe content long before they reach the public with a speed and scale no newsroom could achieve alone. This is how the potential of AI is harnessed, and public service journalists can firmly preserve their position at the centre of editorial decision-making. Within this framework, the EBU has launched the ‘Eurovision News Spotlight’ for fact-checking and open-source intelligence.11 Beyond fact-checking this programme is based on resource-sharing as well as on training and development.
Deeper European Collaboration Is Essential
From a European Public Service Media perspective, technology is not employed to replace human roles but to reinforce them. The investment in skills, training, and editorial awareness will contribute to this end by helping people using new tools with confidence. Greece places emphasis on the evolving discussion about media and AI. Specifically, ERT views journalists, editors, and producers as innovation actors capable of achieving the maximum use of AI. In so doing, ERT considers AI tools as a support mechanism, accelerating routine and mundane tasks and finally allowing journalists to concentrate on more complicated tasks such as investigation, analysis, and storytelling. Our motto is that technology serves the newsroom, and not the other way around.
Truth serves the public good, and the public good needs the protection of strong institutions. Public Service Media such as ERT carry a unique responsibility to remain a reliable point of reference in a period of declining trust in news. Public service media are not in competition for clicks or outrage and are able to escape from the commercialisation trap. They exist to sustain civic dialogue and to strengthen democracy by providing accurate, fair, and verifiable information. That is why deeper European collaboration – through organisations like the EBU – matters. Common standards, shared tools, and a united front to respond to challenges and ensure that technology serves the public interest will make the difference. European synergies will certainly prioritise an ethical framework as AI-assisted content must be disclosed openly. Audiences have the right to know when technology has supported the production or verification of journalism. Transparency is anchored to credibility and vice-versa.
While Public Service Media are exploring the potential of AI tools, the academic debate about their importance is being enriched. A recent article published in The European Journal of Communication proposes an expanded role for Public Service Media along with their core task of providing high quality and verified news and information.12 It argues that they could act as institutions monitoring social media as well as engage in enhancing media literacy. The new debate is an opportunity for both media practitioners and scholars to join hands in support of Public Service Media. Alessandro D’Arma, Christian Fuchs, Minna Horowitz and Klaus Unterberger stress this point and ‘dream of a revitalisation and renewal of Public Service Media.’13
Conclusion
Journalism is a profoundly human enterprise, and public service media are based on values such as independence, truth, and trust. The empowerment of human beings remains a catalyst in the new era of extraordinary technological progress. AI analyses, summarises, translates, and generates content at an astonishing pace. It also processes text, audio, and video in virtually any language. It does not assume responsibility. Neither does it exercise editorial judgment, understand context, ethical consequences, or human impact. This responsibility remains and must remain with journalists and editors. Human verification, human judgment, and clear editorial accountability are essential to ensure that information becomes truth – not just output. AI tools have limits. Reports made by journalists working in the field, the verification of facts, and the publication of stories, which would otherwise go unnoticed, cannot be automated. It is precisely this human presence that transforms data into meaningful, trustworthy journalism.
The essence of journalism – despite the ongoing technological revolution – remains unchanged. It is the relentless pursuit of truth in the service of the public. Technology will continue to evolve. Platforms will come and go. What must not change is the European commitment to the values that define this profession, namely accuracy, fairness, empathy, and accountability. In the end, journalism is not just about information. It is a human act of trust where the ones who report and convey the message and those who watch, listen or read, participate. The strengthening of public service media at EU level is a safe and necessary way forward. Recent results of a study conducted by the EBU and the BBC are alarming. They suggest that AI assistants misrepresent news content 45% of the time, regardless of language or territory.14 Even if the audience does not extensively rely on AI assistants in consuming the news, the alarm bell is already ringing. The acknowledgement of this grim reality will perhaps function as an accelerator for coordinated European responses.
References
1 James Curran, Shanto Iyengar and Inka Salovaara-Moring, ‘Media System, Public Knowledge and Democracy: A Comparative Study’, European Journal of Communication, vol. 24, iss. 1, 1 March 2009 (accessed July 2026).
2 Stephen Cushion, The Democratic Value of News: Why Public Service Media Matter, Bloomsbury Publishing, 2017.
3 Minna Horowit, Icon, Marko Milosavljević and Hildegarde Van den Bulck, ‘The Use of Artificial Intelligence by Public Service Media: Between Advantages and Threats’ in Christo el Morr (ed.) AI and Society: Tensions and Opportunities, Routledge, 2022.
4 EBU website. ‘Generative AI & Public Service Media’, 2024 (accessed July 2026).
5 Marshall McLuhan (with an introduction of Lewis H. Lapham), Understanding Media: The Extensions of Man, The MIT Press, 1964.
6 Reuters Institute, ‘Digital News Report 2026’, 2026 (accessed July 2026)
7 Henrik Rydenfelt, Joel Kiskola and Thomas Olsson, ‘AI in Media Consumption: Charting the Futures of Journalism’, Journalism Studies, vol. 27, iss. 8, 2026 (accessed July 2026).
8 More information on the European Perspective is available on the website: https://www.europeanperspective.net/home (accessed July 2026).
9 European Commission website, ‘Strategic Communication and Countering Foreign Information Manipulation and Interference’, December 2025 (accessed July 2026).
10 Ben Steward, ‘AI Governance Must Put Trust, Information Integrity and Child Safety Front and Centre – Backed by Real Accountability’, 7 July 2026 (accessed July 2026).
11 EBU website, ‘EBU Launches Spotlight Fact-checking Network to Combat Misinformation and Support Trusted News’, 11 April 2025 (accessed July 2026).
12 Annika Sehl, ‘Public Service Media as Pivotal in Combating Misinformation and Disinformation: Prerequisites and Approaches’, European Journal of Communication, vol. 39, iss. 6, 14 November 2024 (accessed July 2026).
13 Alessandro D’Arma, Christian Fuchs, Minna Horowitz and Klaus Unterberger, ‘The Future of Public Service Media and the Internet’, in Christian Fuchs, Klaus Unterberger (eds.), The Public Service Media and Public Service Internet Manifesto, pp. 113-127, University of Westminster Press, 2021.
14 BBC website, ‘Largest Study of Its Kind Shows AI Assistants Misrepresent News Content 45% of the Time – Regardless of Language or Territory’, 22 October 2025 (accessed July 2026).