Deception is as old as conflict itself. Sun Tzu advised warriors to appear unable when able, inactive when active, and far when near. For most of history, the power of such tricks was limited by distance, speed, and the difficulty of reaching large audiences. The digital age shattered those limits. Today, social media platforms, algorithm-driven recommendation systems, and artificial intelligence allow states and political actors to craft messages, spread them globally in seconds, and target specific communities with surgical precision. Information influence is no longer just an accompaniment to military campaigns; it has become an independent weapon of political warfare. Its purpose is often not to persuade people to change their minds, but to exhaust them, polarize societies, undermine faith in institutions, and paralyze democratic decision-making. Russia’s so-called “5D” approach—dismiss, distort, distract, dismay, divide—captures this shift. China and Iran have built their own toolkits, combining social media influence, economic pressure, espionage, and rhetorical manipulation. Romania’s presidential elections at the end of 2024 showed how quickly coordinated foreign interference can shake a democracy. Scholars distinguish between misinformation, which is false but not intentionally harmful; disinformation, which is deliberately deceptive; and malinformation, where true facts are used maliciously. The popular term “fake news” covers only a small part of this spectrum. Authoritarian regimes treat information control as a matter of regime survival, building censored, surveillance-rich digital ecosystems. Democracies face a more difficult question: how to defend against manipulation without destroying free expression, media pluralism, public trust, and political competition. This is not merely a technical or legal challenge. It is a question of governance: which institutions should act, what tasks they should perform, how they should coordinate, and where the line between security and liberty should be drawn. This article explores those questions by looking at fifteen democratic countries and the institutional choices they have made. Rather than offering one ideal blueprint, it shows that democratic responses are shaped by history, geography, strategy, and political culture.
Over the past decade, disinformation has moved decisively from the margins of media studies to the center of national security debates. Politicians, diplomats, and military strategists now talk about hybrid threats, foreign information manipulation, interference, and cognitive warfare. The academic literature reflects this change. A bibliometric search of Scopus found more than 38,000 publications dealing with disinformation and related concepts. Of those, more than 14,000 discussed responses or countermeasures, and nearly 1,200 explicitly linked counter-disinformation to national security. That share more than tripled between 2010 and 2025. The trend is clear: disinformation is increasingly understood as a threat to electoral integrity, social cohesion, and national sovereignty. Yet this securitization has not pushed democracies into a single institutional mold. On the contrary, it has made institutional variation more visible and more important. To understand that variation, the article develops taxonomies of actors and tasks. The actor taxonomy identifies eleven broad groups: executive and national security bodies, ministries, intelligence and defense agencies, specialized counter-disinformation units, cybersecurity authorities, government communication services, electoral and media regulators, media companies and platforms, civil society organizations, academic and research institutions, and international partners. The task taxonomy outlines thirteen groups of work, including monitoring and detection, analysis and attribution, early warning, strategic communication, public awareness, media literacy, regulation, operational response, coordination, research, and foresight. These categories are not fixed boxes; they are analytical lenses. In practice, the same institution may play many roles, and the same task may involve many institutions. What matters is how countries distribute responsibility, create new capacities, and coordinate the whole system. Another crucial insight is that counter-disinformation rarely stands alone as a policy domain. It is usually embedded in broader frameworks: hybrid-threat governance, cybersecurity and digital platform regulation, strategic communication, psychological defense, comprehensive security, or total defense. These “host frameworks” shape which institutions lead, which tasks are prioritized, and how coordination works.
From the taxonomies and policy reviews, the article constructs six broad institutional models. The first is distributed governmental coordination. Here, responsibilities remain spread across ministries and agencies, and coordination happens through existing structures like national security councils, cabinet offices, and interagency committees. This model preserves institutional continuity and avoids building new bureaucracies. The second is specialized interagency coordination. A dedicated council, commission, or task force brings senior officials together and is often supported by a permanent secretariat. This improves coherence without creating a large operational agency. The third model is a dedicated counter-disinformation agency. A single body concentrates expertise, responsibility, and public visibility. This can be effective, but it also carries risks: over-centralization, political pressure, and rivalry with existing institutions. The fourth is the embedded-function model. Counter-disinformation is placed inside an institution with a broader mandate, such as a cybersecurity agency, a hybrid-threat center, a psychological defense organization, or a strategic communication department. This uses existing capacity and can be less disruptive. The fifth is the public-private partnership model. Government, platforms, media, academia, and civil society work together through structured arrangements, reflecting the fact that the information environment is an ecosystem. The sixth is the externally implemented or distributed societal model. Here, non-governmental actors do most of the work—fact-checkers, civic-tech organizations, universities, media councils, and volunteer networks—while the state plays a limited or indirect role. These models are not mutually exclusive. Countries frequently combine elements. They are best understood as reference points in a wider space of possibilities, not boxes on a flowchart. For example, placing a unit under a national security council is different from creating a free-standing agency; designing a voluntary platform code of conduct is different from imposing legal duties. The models help policymakers see what options exist and why particular choices may be more or less compatible with democratic governance.
To describe the differences between countries more precisely, the article proposes five dimensions of governance. The first is centralization versus decentralization. Some systems concentrate coordination and decision-making at the center; others distribute it widely across public and private actors. The second is specialization versus distributed responsibility. Some countries create dedicated agencies or units; others fold counter-disinformation into existing ministries, communication offices, cybersecurity bodies, or regulatory agencies. The third dimension concerns the participation of non-state actors. Is cooperation with platforms, media, and civil society legally required, co-regulated, or entirely voluntary? Democracies cannot respond effectively without private actors, but the degree of compulsion varies enormously. Fourth is enforcement-oriented versus standards- and resilience-oriented approaches. Some systems rely on laws, fines, content restrictions, takedowns, and sanctions. Others prefer professional standards, voluntary codes, transparency, media literacy, and public communication. Fifth is the framing itself: security-centric versus resilience-centric. Some countries treat disinformation primarily as a foreign interference or hybrid-threat challenge, demanding intelligence-led responses. Others emphasize societal preparedness, public trust, and democratic robustness. These dimensions are related, but they are not the same. A country can be highly specialized without being coercive; it can be security-conscious and still rely heavily on voluntary civic action. The dimensions also reveal trade-offs. Centralization can improve coordination and accountability but may reduce flexibility and local initiative. Decentralization can mobilize broad energy and creativity but requires trust and effective coordination. Mandatory obligations on platforms and media may increase accountability but risk chilling legitimate speech. Enforcement can remove harmful content quickly but cannot replace the slow work of building media literacy, public trust, and institutional resilience. The goal is not to choose one end of every spectrum forever; it is to design a system that balances these tensions in a way that fits national context and democratic values.
When real countries are placed in this space, the diversity becomes vivid. Finland is the archetype of distributed societal resilience. It has no central counter-disinformation agency; instead, comprehensive security unites authorities, business, organizations, and citizens around the protection of vital social functions. Sweden embeds the work in psychological defense, with a dedicated agency that identifies and counters undue information influence while safeguarding free opinion formation. Norway follows a similar path through total defense and civil-military preparedness. Estonia, Lithuania, and Latvia occupy a more security-conscious zone. Their geopolitical exposure to Russian influence campaigns pushes them toward hybrid-threat governance, strategic communication, and stronger coordination, though they remain institutionally plural. Poland and Czechia combine coordination mechanisms with existing institutions rather than creating large new bureaucracies. Romania is still adapting, moving from fragmentation toward more structured coordination after the shock of the 2024 election. France represents centralized state coordination: the General Secretariat for Defence and National Security and the VIGINUM agency form a coherent system tied to information sovereignty. Ukraine is the most wartime-operational case. Its Center for Countering Disinformation sits under the National Security and Defence Council, security services actively dismantle bot farms, and the Ministry of Culture coordinates strategic communication. Yet Ukraine also relies heavily on civil society, from StopFake to volunteer bloggers and professors, making it a hybrid of securitization and whole-of-society resistance. Taiwan is a striking outlier. Under sustained pressure from Chinese influence operations, it has chosen low coercion and high civic engagement, building digital immunity through transparency, participatory governance, and civic-tech communities rather than censorship. Israel integrates information governance with national security but depends largely on voluntary and private initiatives. The United Kingdom is distributed but operationally coordinated, using the RESIST framework and a cross-government Defending Democracy Taskforce. The United States is the most plural and politically contested case: capabilities are spread across federal agencies, cybersecurity institutions, election officials, platforms, and civil society, but polarization and constitutional protections make centralized responses difficult. These profiles show that threat severity does not mechanically determine institutional design.
Taken together, the central lesson is that democratic counter-disinformation governance is a matter of fit, not formula. There is no universal model that every country should copy. Securitization—treating disinformation as a national security issue—does not automatically lead to centralized, coercive systems. Finland and Taiwan demonstrate that high threat awareness can coexist with distributed, resilience-oriented governance. France and Ukraine show that more security-focused models can be legitimate and effective in their contexts. Institutional choices reflect strategic culture, legal tradition, administrative heritage, trust in the state, and the nature of the threat. Still, some general principles emerge. Clarity about roles matters, whether responsibilities are concentrated or dispersed. Coordination must be real, not just formal; councils and task forces need resources, access, and authority. Legitimacy is a strategic asset. Overly intrusive measures can hand adversaries a propaganda victory and undermine the very trust they are meant to protect. Resilience-building is not a soft option; media literacy, public awareness, and transparent communication are operational necessities, especially in acute crises. Finally, no single actor can do everything. Platforms, media, academia, and civil society are not stakeholders to be managed; they are part of the defense ecosystem itself. The article has limitations. It relies mostly on public documents and formal mandates, which cannot capture informal practices, classified operations, or implementation gaps. Its country scores are qualitative judgments, not precise measurements. The field is evolving rapidly. Generative artificial intelligence, new platform regulations, and shifting threat perceptions will continue to reshape national architectures. Future research should examine implementation in practice, study institutional trajectories over time, and analyze how democratic systems adapt to technological change without imitating authoritarian models. The ultimate goal is not to build a perfect machinery of control. It is to preserve an open society that can defend itself without becoming something other than itself.