Democracy Without Borders? Global Citizenship Between Platform Power, AI and the Crisis of Knowledge

Is the idea of global democracy over? Or has the time now come for more realism, following the democratic gains of recent decades? Ten years after Ingolf Pernice’s visionary speech at the Congreso del Futuro in Santiago de Chile, it is worth looking back, and above all forward. What became of the idea of governing the 21st century democratically? And what new questions do platform economics, artificial intelligence and the role of knowledge raise for democracy today?

Under what conditions can we understand ourselves as world citizens, and what would it take to meet global challenges democratically? Ingolf Pernice, professor of constitutional law at Humboldt University of Berlin and one of the founding directors of the Alexander von Humboldt Institute for Internet and Society (HIIG), posed this question in January 2016 at the Congreso del Futuro in Santiago de Chile.* His answer was as ambitious as it was systematic: democracy must not end at the state border. We need constitutionalism, understood as a striving for human-centredness of law and laws at every level: the individual, the local community, the region, the state and global governance.

Democracy Beyond the Nation State

To this end, Pernice developed the model of multilevel constitutionalism. The term sounds unwieldy at first, but it describes something simple. We tend to think of constitutions in national terms. Germany has its Basic Law, Austria its Federal Constitutional Act, the United States their Constitution. These documents set out the rights citizens hold and how power is organised and limited. Pernice’s argument is that this is no longer enough. Many of the challenges we face can no longer be resolved nationally, so we need structures at other levels that limit power and grant it democratic legitimacy. The European Union is the best-known example. It has its own rules, its own institutions and its own fundamental rights, which can take precedence over national law.

What distinguishes Pernice’s approach is that he builds this idea consistently from the individual outward. For Pernice, democracy begins with the individual citizen, who is the true sovereign within a democratically organised system. All political power must be traceable back to that individual will. Institutions do not govern by their own right, but because people have conferred that right upon them.

Staying Capable of Global Action

When problems such as climate change, financial markets or cybersecurity exceed the capacity of individual states to act, democratically legitimated institutions are needed beyond the state. Not a world state, which would only distance citizens further from decision-making, but an additional, complementary layer: a global-democratic layer that ensures societies remain capable of acting where national borders end, always organised according to the principle of subsidiarity. This principle holds that decisions should always be taken as close to the people affected as possible. Only what cannot be resolved locally or nationally moves up a level.

This is where the internet comes in. For Pernice, the internet is an instrument that makes global democracy realistic for the first time. It is the first means of communication to connect people worldwide, rapidly and in networks (the fax machine, itself now nearly two centuries old, brought the world closer together too, but it remained a point-to-point connection). The internet, by contrast, creates the conditions under which people can deliberate and decide democratically across borders. How it is regulated, and by what rules, is therefore not a technical question but a deeply political one, since these rules also govern the internet’s emancipatory potential for communication.

Ten Years On: Vision Meets Reality

Ten years later, Pernice’s idea has lost none of its force. If anything, the opposite is true. When he spoke in 2016, platforms could still set their own rules largely unchecked. The world of 2026 looks very different. The model of multilevel constitutionalism has been confirmed by the development of European digital law. The EU’s Digital Services Act and Digital Markets Act set out how digital platforms such as Meta, TikTok, Google, and Amazon may handle content and exercise market power, since these companies shape, through their business models and technical infrastructure, what becomes visible in public discourse and how we communicate with one another. The AI Act determines, across Europe, which AI applications are permitted and under what conditions. The United Nations Global Digital Compact seeks to anchor global principles for an open and secure digital space. What was formulated as a forward-looking vision ten years ago has since taken on concrete shape.

Platforms, AI, Knowledge: Three New Dimensions of Global Governance

At the same time, the conditions for democratic self-determination in the digital sphere have changed fundamentally since 2016. Platform power, algorithmic curation and the question of what role scientifically produced knowledge plays in democratic discourse all challenge the framework Pernice sketched in 2016. Self-determination presupposes that people can decide for themselves what they see, what they believe and how they participate politically. Pernice envisaged the internet as a space in which exactly this would be possible. That hope was not misplaced, but realising it has proven more complicated. Over the past decade, at least three fields have opened up that extend the model in important ways.

1: Platforms as New Centres of Power

In 2016, the debate over the power of digital platforms was only beginning. Today it is clear that a handful of companies shape the infrastructure of public discourse to a considerable degree. Research from HIIG’s now-concluded research group, Global Constitutionalism and the Internet, shows that platforms must be understood as transnational sites of power interwoven with national constitutional orders (Kettemann, 2024). This power crosses state borders. A decision by Meta or X affects people in Germany, Brazil and Kenya simultaneously, without any national parliament having voted on it. Pernice’s constitutional thinking, grounded in the individual, offers a compass for bringing these levels, too, under democratic control.

One concrete proposal is the establishment of platform councils (Kettemann & Schulz, 2023). The underlying idea is that platforms today make decisions about freedom of expression, visibility and exclusion that courts or parliaments once made. Platform councils would be democratically legitimated oversight bodies, involving users, civil society and experts, tasked with scrutinising platform power. They would democratise platform governance without making the state the sole arbiter.

2: Digital Participation and Its Limits

Pernice’s vision relied on e-voting, digital deliberation and the possibility of a global public sphere. The underlying idea was that if people could not only vote online but also weigh arguments and deliberate together, democratic participation beyond national borders would become possible for the first time. These tools exist and continue to develop. But the environment in which they operate has changed. Platforms today use algorithms to determine which content is visible and which is not. They earn money through attention, not through the quality of democratic discourse. Such decisions carry democratic consequences and are therefore not technical side effects but structural changes to the infrastructure on which democratic participation depends. The question is no longer whether the internet can enable democratic participation at the global level, but how. Equally, we must ask what protective mechanisms the internet’s democratic infrastructure requires.

European platform law offers one example. In a joint commentary on the DSA and DMA, Mast and colleagues examined the role European platform law can play for democratic communication (Mast, Kettemann, Dreyer & Schulz, 2024). A key finding is that the DSA obliges platforms to disclose their effects on public discourse, since lawmakers can only act on an informed basis once it is known how algorithms shape democratic debate. Translating platform data into political decision-making knowledge in this way marks an important step. HIIG’s project Human in the Loop? shows what this can look like in practice. It examines how the collaboration between humans and machines in content moderation on social media platforms can be designed so that democratic goals are not lost along the way, including enabling respectful and safe online discourse through the detection and containment of hate speech, insults or illegal content, while preserving freedom of expression and keeping moderation decisions accountable to all.

Both projects are part of a larger effort to democratically shape the emerging “normative order of the internet” (Kettemann, 2020). What remains open is how this order can be designed so that it not only permits democratic negotiation within society but actively protects it.

3: Knowledge as a Democratic Resource

Pernice argued in 2016 that global democracy rests on access to information, education and a functioning public discourse. Democratic self-determination requires a shared basis of knowledge, which in turn requires that people hold shared assumptions about which information, facts and opinions can be considered reliable, whether scientific studies, independent journalism or expert assessment. This foundation is under pressure. Growing inequality, political polarisation and declining trust in institutions such as parliaments and courts, as well as in legally secured procedures such as elections and legislation, all play a part, alongside many other factors. Digital platforms reinforce these developments, since content that provokes reaction is algorithmically favoured for distribution regardless of its reliability. New technologies also lower the barriers to producing content at scale whose reliability is difficult to assess. Where shared facts once formed the basis of political debate, different groups increasingly accept their own separate sets of information as true.

When facts are contested, when scientific expertise comes under pressure and when algorithmic systems steer the flow of information, the question of how knowledge enters democratic processes becomes itself a constitutional, and constitutionalising, question (Mansell et al., 2025). At the local political level, scientific knowledge is increasingly displaced by commercial consultancies that present themselves as scientific without undergoing academic quality assurance (Fecher, Sokolovska & Wagner, 2025). At the same time, hostility can lead researchers to avoid controversial topics or withdraw from public debate altogether (Blümel et al., 2026). Even the rules governing which knowledge should feed into democratic processes remain a continuous negotiation between states, platforms and civil society, rather than a settled framework (Kettemann & Tiedeke, 2025).

What Remains to Be Done

Pernice’s speech ended with a sentence that remains as relevant today as it was in 2016: “If we learn to take ourselves seriously as the owners of our political system, it is the internet that allows us to establish digital democracy at the global level and so to govern ourselves as global citizens, the 21th century [sic]” (Pernice, 2016).

This insight, that democracy cannot end at the borders of the nation-state and that all of us must develop a sense of ownership over the structures in which our future is negotiated, grows clearer and more urgent with every crisis, be it pandemic, climate emergency or the geopolitical fragmentation of the digital sphere. The central challenges Pernice identified in 2016 can be distilled into two questions for the future.

1: How can the concept of global citizenship be developed further so that it can unfold its full potential in an age of platform-based governance?

People today live not only within state-regulated orders but also within digital spaces whose rules are set by private platform companies, with no democratically elected institutions having any say (Wagner et al., 2025). The individual as sovereign, Pernice’s starting point, must therefore be thought further. Not only as a citizen of a state, but also as a user of platforms whose rules increasingly shape everyday life. In practice, the multilevel model of shared public responsibility that Pernice developed has taken the form of a mosaic of different approaches, in which horizontal and vertical power structures and regulatory mechanisms are inseparably interlinked. 

At the global level, the Internet Governance Forum, successful for twenty years now, at least offers a space in which states, civil society and companies can deliberate together on the fundamental rules of internet-based communication. For new technologies such as quantum computing, a similar format has already been proposed, the so-called Quantum Governance Forum (COMEST, 2025).

2: What role does scientifically produced knowledge play in democratic processes at the global level? And how can the institutional conditions for it be strengthened?

Technologically and legally, the question is who sets the rules for new technologies and how these rules gain democratic legitimacy. Socially, the question is how a shared, reliable basis of knowledge can still emerge in a digitalised world, one on which democratic debate can be built and shaped. Both dimensions are inseparable and must be studied from an interdisciplinary perspective; at HIIG, this work is carried out within the research foci of New Technologies and the Future of Law and Democratic Change and Knowledge.

The Future of Global Democracy

The goal remains a democratic order in which the internet enables more participation while also improving the quality of collective decisions. The vision Ingolf Pernice put forward in 2016 remains unrealised today, and that is precisely what makes it worth pursuing. The technical possibilities for achieving it are further advanced in 2026 than ever before. Governing the 21st century also means securing the foundations of knowledge and the communicative orders that make informed democratic decisions possible in the first place. What began as a vision is now a concrete task, for researchers, lawmakers and civil society alike.

Notes

* An elaborated version of Ingolf Pernice’s speech at the Congreso del Futuro was subsequently published, and can be read, in the Internet Policy Review (Pernice, 2016).

References

Blümel, C., Fähnrich, B., Fecher, B., Sokolovska, N., & Brandt, E. (2026). Understanding hostility towards researchers. A collection of empirical insights into science hostility and attacks on the integrity of researchers. In Tackling hostility towards scientists. Five complementary educational resources to equip institutions and individual researchers with knowledge and skills (pp. 6–11). Alexander von Humboldt Institute for Internet and Society. https://doi.org/10.5281/zenodo.19468717

COMEST (2025). The ethics of quantum computing. World Commission on the Ethics of Scientific Knowledge and Technology. https://unesdoc.unesco.org/ark:/48223/pf0000398114

Fecher, B., Sokolovska, N. & Wagner, G. G. (2025). Exploring scientific advice for local politics in Germany: the critical role of quasi-scientific actors. Science and Public Policy, 1–12. https://academic.oup.com/spp/advance-article/doi/10.1093/scipol/scaf070/8328260

Kettemann, M. C. (2020). The Normative Order of the Internet. A Theory of Rule and Regulation Online. Oxford: Oxford University Press. https://global.oup.com/academic/product/the-normative-order-of-the-internet-9780198865995?cc=de&lang=en&

Kettemann, Matthias C.; Schulz, Wolfgang (eds.) (2023). Platform://Democracy – Perspectives on Platform Power, Public Values and the Potential of Social Media Councils. Hamburg: Verlag Hans-Bredow-Institut. https://www.ssoar.info/ssoar/handle/document/86524

Kettemann, M. C. (2024). Die Mächtigen regulieren: Das Potenzial von Plattformräten. Digital Society Blog. https://www.hiig.de/publication/die-maechtigen-regulieren-das-potenzial-von-plattformraeten/

Kettemann, M. C. & Tiedeke, A. S. (2025). Normativities, Normative Orders, and Pluralism. In De Gregorio, G., Pollicino, O. & Valcke, P. (Eds.), The Oxford Handbook of Digital Constitutionalism. Oxford: Oxford University Press. https://academic.oup.com/edited-volume/58210/chapter-abstract/504137157?login=false&redirectedFrom=fulltext

Mansell, R., Durach, F., Kettemann, M. C. et al. (2025). Information Ecosystems and Troubled Democracy: The State of Knowledge on News Media, AI, and Data Governance. Gothenburg: Nordicom. https://www.nordicom.gu.se/en/publications/information-ecosystems-and-troubled-democracy

Mast, T., Kettemann, M. C., Dreyer, S. & Schulz, W. (Eds.) (2024). Digital Services Act / Digital Markets Act: Kommentar. München: C.H. Beck. https://www.beck-shop.de/mast-kettemann-dreyer-schulz-dsa-dma/product/34660327?srsltid=AfmBOopDCGXFZXAlI_6bVEJG0w4KU92ktbkdNmXgBLM3sIQltKU48P1P

Pernice, I. (2016, 4 May). Multilevel constitutionalism and e-democracy. Internet Policy Review. https://policyreview.info/articles/news/multilevel-constitutionalism-and-e-democracy/407

Wagner, B., Kettemann, M. C., Vieth-Ditlmann, K. & Montgomery, S. (Eds.) (2025). Research Handbook on Human Rights and Digital Technology. Cheltenham: Edward Elgar. https://www.e-elgar.com/shop/gbp/research-handbook-on-human-rights-and-digital-technology-9781035308507.html?srsltid=AfmBOopSnJNbgnNiX7RBXTfly6A6GisXW_My_weChors_2iWTZ9Nr1_F