Algorithmic content recommendation is widely regarded as an unavoidable feature of today’s digital media environment, particularly as a response to information overload. A new report from the Digital Democracy Centre at the University of Southern Denmark, however, points to growing concern among media policy and regulatory experts across the Nordic–Baltic region about how major online platforms currently implement these systems. The report is the result of research conducted as part of the DDC × TrygFonden Fellowship Programme, and prepared by Fellow researcher Guna Spurava, PhD, & Professor Arjen van Dalen.
The report, Algorithmic Content Recommendation: Insights from Nordic–Baltic Media Policy and Regulatory Stakeholders, draws on in-depth interviews with 14 experts from eight countries. It examines how those working with media policy and regulation assess the benefits, risks, and broader societal implications of algorithmic recommendation systems used by platforms such as Facebook, YouTube, TikTok, X, and Google.
Experts consistently describe algorithmic recommendation as “a core component of global platforms’ business models”, designed to “maximise user engagement, prolong user activity on platforms, and convert attention into revenue”, the report states.
While interviewees acknowledge that such systems can help users navigate large volumes of information, many emphasise that the engagement-driven logic shaping current implementations has significant consequences for society.
One of the central concerns identified in the study relates to democratic discourse.
– Media policy and regulatory actors warn that algorithmic systems increasingly shape public discourse through opaque processes of content selection and amplification, influencing which issues gain visibility and how public debate is structured, notes Spurava.
Several respondents point to the risk of increased polarisation and fragmentation of the public sphere, particularly in smaller Nordic–Baltic states where information environments may be more vulnerable to targeted political messaging and foreign influence.
Beyond democratic processes, the report highlights challenges to the integrity of the information ecosystem, including reduced diversity of content exposure, the spread of misinformation and disinformation, and economic pressure on legacy media.
At the individual level, stakeholders raise concerns about user agency and access to information. As van Dalen observes, several respondents view opaque practices of data collection, profiling, and content filtering as posing fundamental challenges to freedom of information and human rights in algorithmically curated media environments.
Rather than pointing to a single remedy, the report emphasises the need for a comprehensive and multi-layered response.
Strong EU-level regulation – particularly through frameworks such as the Digital Services Act – is widely seen as necessary, but insufficient on its own. The interviewees hightlight the importance of complementary measures, including public awareness-raising, media and algorithmic literacy, alternative technological and design approaches, and sustained support for independent, editorial journalism as foundations for a resilient democratic digital ecosystem.