News Wars: Media, Disinformation and Electoral Integrity in Europe

• Introduction
• Chapter 1 Mapping Europe’s Information Battlefield
• Chapter 2 Actors and Incentives: States, Parties, Proxies, and Profiteers
• Chapter 3 Platforms as Battlegrounds: Algorithms, Ads, and Amplification
• Chapter 4 Tactical Playbooks I: Seeding, Flooding, and Framing
• Chapter 5 Tactical Playbooks II: Forgeries, Deepfakes, and Synthetic Networks
• Chapter 6 Cross-Border Operations and Diaspora Media
• Chapter 7 Domestic Manipulation and Campaign Dirty Tricks
• Chapter 8 Local News Deserts, Capture, and the Economics of Attention
• Chapter 9 Hack-and-Leak: From Breach to Narrative
• Chapter 10 Memes, Micro‑Influencers, and Networked Propaganda
• Chapter 11 Messaging Apps, Encrypted Channels, and Closed Communities
• Chapter 12 Language, Identity, and Minority Audiences
• Chapter 13 Election Timelines: Vulnerable Windows and Critical Moments
• Chapter 14 Verification Methods: OSINT, Forensic Media, and Fieldwork
• Chapter 15 Network Mapping: Graphs, Clusters, and Influence Flows
• Chapter 16 Attribution and Accountability: What We Can and Cannot Know
• Chapter 17 Measuring Impact: Surveys, Experiments, and Behavioral Signals
• Chapter 18 Platform Policies: Gaps, Evasions, and Enforcement
• Chapter 19 Regulation in Practice: EU DSA, DMA, GDPR, and Member-State Laws
• Chapter 20 Public Service Media, Fact-Checking, and Cross-Border Collaboratives
• Chapter 21 Newsroom Resilience: Protocols, Workflows, and Rapid Response
• Chapter 22 Election Administration: Risk, Communication, and Crisis Playbooks
• Chapter 23 Rebuilding Trust: Transparency, Corrections, and Community Engagement
• Chapter 24 Prebunking and Education: Media Literacy at Scale
• Chapter 25 What’s Next: AI-Driven Operations and Defensive Innovation

Elections are moments when democracies tell the truth about themselves. They are also moments when adversaries—foreign and domestic—work hardest to bend that truth. Across Europe, information operations exploit social divisions, weaken trust in media and institutions, and attempt to shape outcomes at precisely the times when citizens need clarity most. This book is a practical, investigative account of how those operations work, why they succeed, and how journalists, regulators, platforms, and election officials can counter them without compromising democratic values.

The pages that follow map the sources, networks, and tactical playbooks that drive contemporary disinformation. Rather than treating manipulation as a single problem with a single fix, we examine layered systems: from opportunistic grifters seeking clicks and ad revenue, to coordinated political actors, to state-linked operations that blend espionage with public influence. We look at how narratives are seeded, how fringe claims are laundered into mainstream discourse, and how amplification engines—recommendation algorithms, influencer networks, ads, and messaging apps—turn sparks into wildfires.

Our approach is deliberately hands-on. We present verification methodologies that reporters and analysts can apply immediately: open-source intelligence techniques, forensic media checks for images, audio, and video, and network mapping to uncover coordinated inauthentic behavior. Readers will learn how to trace claims back to original sources, recognize manipulation cues, and document evidence in ways that stand up to editorial scrutiny and, when necessary, legal challenge. These methods are paired with field-tested workflows that help newsrooms and election teams move quickly without sacrificing accuracy.

Because information operations do not respect borders, we place European elections within a cross-border context. Language communities, diaspora media, and transnational platforms allow narratives to leap jurisdictions in minutes. We explore how disinformation adapts to local histories and cultural identities, and why minority and multilingual audiences often face tailored manipulation. Case studies illustrate how “hack-and-leak” campaigns, synthetic personas, deepfakes, and micro-targeted ads converge around critical moments in the election calendar—from candidate registration and early voting to debate nights and result announcements.

Defense requires more than detection. We assess platform policies and enforcement practices, highlighting gaps that adversaries exploit and documenting evasive techniques that evolve in response. We examine the regulatory landscape—including the Digital Services Act, the Digital Markets Act, data protection frameworks, and relevant member-state laws—to clarify where accountability mechanisms exist, when they are effective, and how they can be abused. Regulatory tools can reduce systemic risk, but they must be paired with institutional capacity, independent oversight, and robust civil-society engagement.

Resilience is the throughline of this book. For media outlets, that means editorial protocols for suspected manipulation, transparent corrections, and audience engagement that builds trust before a crisis hits. For election authorities, it means risk assessment, clear communication channels, scenario planning, and “prebunking” campaigns that inoculate the public against predictable falsehoods. We offer templates for rapid response, checklists for cross-functional coordination, and metrics to evaluate whether interventions are working.

Finally, we look ahead. Artificial intelligence accelerates both offense and defense: models can fabricate persuasive content at scale, but they can also help detect coordination, authenticate media, and forecast where manipulation is likely to strike next. Europe’s democratic health will depend on whether institutions, platforms, and the public can adapt as quickly as adversaries do. This book equips practitioners with the tools to meet that challenge: to identify threats early, respond with precision, and strengthen the informational commons on which free and fair elections depend.

Europe, a continent of ancient borders and modern democracies, finds itself increasingly contested not by armies, but by algorithms and narratives. The battlefield isn't always visible; it’s often a subtle contest for attention and belief, waged across social feeds, encrypted chats, and niche news sites. Understanding this landscape requires moving beyond simplistic notions of "fake news" and instead mapping the complex ecosystems where information, and misinformation, flourish and intertwine. This chapter will lay the groundwork, identifying the key features of Europe's diverse media environments and pinpointing the vulnerabilities that adversaries exploit.

The European information space is a mosaic of languages, cultures, and historical experiences. Unlike the more homogeneous media landscapes found in some other parts of the world, Europe’s linguistic diversity alone presents a formidable challenge. Every language acts as a potential vector for tailored narratives, often insulated from scrutiny by a broader public or international fact-checking efforts. A disinformation campaign targeting Slovakian speakers might be entirely invisible to a French journalist, even if both reside within the same regulatory framework. This fragmentation creates fertile ground for localized influence operations that can fly under the radar of pan-European analysis.

Consider, for instance, the varying levels of trust in traditional media across the continent. In some Northern European countries, public service broadcasters and established newspapers still command relatively high levels of public confidence. Citizens there might be more inclined to dismiss sensational or unsubstantiated claims. Conversely, in parts of Southern and Eastern Europe, historical experiences with state-controlled media or pervasive political partisanship have eroded faith in mainstream outlets. Here, alternative news sources, often highly biased or outright fabricated, can gain traction more easily, tapping into existing cynicism and grievances. This pre-existing trust deficit acts like a weakened immune system, making populations more susceptible to the next viral falsehood.

The political spectrum itself contributes to the complexity. Europe is home to a robust array of political ideologies, from centrist consensus to the far-left and far-right fringes. Each segment of this spectrum often has its own preferred media outlets, its own echo chambers, and its own narrative vulnerabilities. A story designed to inflame anti-immigrant sentiment, for example, might find immediate resonance within certain nationalist online communities, regardless of its factual basis. Conversely, a narrative critical of established institutions might find a receptive audience among anti-establishment groups. The battle, therefore, is not just for a universal "truth," but for the perceived truth within these ideologically siloed communities.

Technological adoption rates and digital literacy levels also carve out distinct features on this information map. While broadband penetration is generally high across the EU, significant disparities exist in how citizens engage with digital platforms and critically assess online content. Older demographics, for example, might be less adept at identifying manipulated images or distinguishing between legitimate news sites and propaganda outlets. Younger generations, while digitally native, are not immune; they often consume news through social media feeds curated by algorithms that prioritize engagement over accuracy, making them susceptible to emotionally charged or polarizing content. These varying levels of media literacy mean that a single disinformation campaign might require different tactics to succeed in different national or demographic contexts.

Moreover, the shadow of historical grievances and geopolitical tensions looms large over Europe's information space. Memories of past conflicts, occupations, and ideological divisions are easily weaponized. Narratives that exploit these historical wounds can be incredibly potent, even if factually dubious. For example, disinformation campaigns targeting the Baltic states or Poland often play on anxieties related to Russia, echoing Soviet-era propaganda or attempting to rewrite historical events. Similarly, narratives around migration can tap into historical fears of cultural dilution or economic strain, regardless of current realities. These deep-seated emotional triggers are a goldmine for information manipulators.

The legal and regulatory frameworks, while increasingly harmonized at the EU level, still present a patchwork of approaches at the national level. The Digital Services Act (DSA) and Digital Markets Act (DMA) represent significant steps towards a more unified approach to platform accountability, but national laws on defamation, hate speech, and media ownership continue to vary. This regulatory labyrinth can create opportunities for actors to exploit jurisdictional loopholes, launching campaigns from countries with more permissive speech laws or less robust enforcement mechanisms. The cross-border nature of information flow often outpaces the ability of national legal systems to respond effectively.

Economic factors also shape the battlefield. The decline of traditional advertising revenues has hit many European news organizations hard, particularly at the local level. This has led to news deserts—areas where independent, professional journalism has diminished or disappeared entirely. These voids are quickly filled by alternative sources, some legitimate, many not. Without reliable local news to hold power to account and provide factual information, communities become more vulnerable to rumors, local political manipulation, and the narratives pushed by well-funded external actors. The economics of attention, where clicks and engagement drive revenue, further incentivizes sensationalism and polarizing content, even from otherwise legitimate sources.

The sheer volume of content produced daily further complicates matters. The "firehose of falsehood" approach, where adversaries flood the information environment with so much conflicting and contradictory information that the public becomes overwhelmed and disengaged, is a common tactic. It’s not just about convincing people of a lie, but about fostering a general sense of confusion and distrust in any information source. When citizens can no longer discern truth from fiction, they may disengage from the democratic process entirely, which is often the ultimate goal of those seeking to destabilize democracies.

Finally, the increasing sophistication of information operations means that the battlefield is constantly evolving. What began with simple fabrication and amplification has progressed to include deepfakes, synthetic media, and AI-generated content that blurs the lines between reality and simulation. The cat-and-mouse game between those who spread disinformation and those who try to counter it is relentless. As new technologies emerge, so do new vulnerabilities and new tactical playbooks. Remaining effective in this environment requires continuous learning, adaptation, and a deep understanding of the diverse and dynamic European information landscape. This chapter, therefore, serves as the essential primer for navigating the complexities we will explore in the subsequent pages.

The information battlefield in Europe isn't a free-for-all; it's a meticulously orchestrated, if often chaotic, arena where a diverse cast of characters pursues an equally diverse set of goals. Understanding who is doing the manipulating and why is paramount to developing effective countermeasures. This chapter peels back the layers to expose the key actors, from the visible political parties to the shadowy state-sponsored operations and the purely mercenary profiteers, examining the incentives that fuel their forays into disinformation.

At the apex of sophistication and strategic intent are often nation-states. These actors engage in information operations not for clicks or cash, but for geopolitical advantage. Their objectives range from undermining democratic institutions in rival nations, destabilizing alliances, swaying public opinion on specific foreign policy issues, or simply fostering a general sense of distrust and confusion that weakens an adversary from within. Russia, for example, has been a prominent actor in this space, employing a sophisticated blend of overt propaganda, covert influence operations, and cyber warfare to achieve its foreign policy aims across Europe. Their tactics frequently involve exploiting existing social cleavages, amplifying extremist voices, and disseminating narratives that align with their strategic interests, often without direct attribution.

These state-level operations are rarely monolithic. They often involve a complex web of government agencies, intelligence services, state-funded media outlets, and a network of proxy organizations or individuals. The lines between these entities can be deliberately blurred, making attribution a painstaking process. For instance, a state might fund a seemingly independent think tank that then publishes research echoing state narratives, or it might subtly support online media outlets that promote its geopolitical agenda. The goal is to create a plausible deniability, allowing the state to achieve its objectives while minimizing direct accountability.

Beyond state actors, domestic political parties and their affiliates are significant players in the disinformation landscape, particularly during election cycles. Their incentives are straightforward: to win votes, damage opponents, and control the political narrative. While outright falsehoods are a common tactic, political disinformation often operates in a grey area, bending facts, taking quotes out of context, or crafting highly misleading interpretations of events. The rise of hyper-partisan media, both traditional and online, provides a ready-made echo chamber for these narratives, allowing them to spread rapidly among sympathetic voters.

Political parties may directly employ digital strategists or public relations firms to craft and disseminate misleading content. They might also leverage party youth wings, volunteer networks, or sympathetic online communities to amplify their messages. The incentive here isn't necessarily to completely deceive the public, but to energize their base, demoralize the opposition, and shape the public discourse in a way that favors their electoral prospects. The speed and reach of social media platforms have supercharged these domestic influence campaigns, making it easier for parties to bypass traditional media gatekeepers and speak directly to their target audiences, often with highly tailored messages.

Then there are the "proxies" – a broad category encompassing various groups and individuals who, wittingly or unwittingly, serve the interests of larger actors. These can include ideologically aligned activists, online trolls, or even seemingly legitimate news organizations that have been co-opted or founded with the specific purpose of promoting a particular agenda. For state actors, proxies offer a crucial layer of deniability. A foreign government might fund a website that appears to be a local news source, but which consistently publishes articles critical of the incumbent government or promotes narratives beneficial to the sponsoring state. This creates the illusion of organic, local discontent, rather than foreign interference.

The motivations of these proxies can vary widely. Some are ideologically committed to the cause they promote, genuinely believing in the narratives they spread. Others might be mercenary, paid to disseminate specific content or amplify particular messages. In some cases, individuals might become unwitting proxies, sharing misleading content without realizing its true origin or intent, simply because it aligns with their existing biases or resonates emotionally. Identifying and disentangling these proxy networks is a significant challenge for investigators, as they often rely on fake accounts, burner phones, and encrypted communication to obscure their true identities and affiliations.

A distinct, yet equally impactful, category of actors are the "profiteers." These individuals or groups are driven primarily by financial gain, often with little to no ideological allegiance. Their business model revolves around generating clicks, views, and engagement, which translates into advertising revenue. They create sensational, often false or highly misleading content, knowing that such material tends to go viral on social media platforms. The content itself might be politically charged, but the motivation behind its creation isn't political; it's purely economic.

These profiteers often operate from obscure locations, setting up networks of sham websites, social media pages, and fake accounts designed to mimic legitimate news sources or influential personalities. They will churn out articles designed to provoke strong emotional responses – anger, fear, outrage – as these emotions are highly effective at driving sharing and engagement. While their primary goal is financial, their activities have a corrosive effect on the information environment, flooding it with unreliable content and contributing to the overall erosion of trust in media. Their willingness to publish anything for a profit makes them particularly dangerous, as they can inadvertently or deliberately amplify narratives originating from state or political actors, blurring the lines even further.

The incentives for each of these actor types often intertwine. A state actor might commission a disinformation campaign, which is then amplified by a network of ideologically aligned proxies, and further spread by profiteers who see an opportunity to generate traffic from the sensational content. This creates a complex ecosystem where discerning the original intent and ultimate beneficiary of a particular piece of disinformation becomes incredibly difficult. It’s rarely a single actor operating in isolation, but rather a confluence of interests and tactics.

Understanding the why behind the manipulation is crucial for effective counter-strategies. If the incentive is geopolitical destabilization, then the response needs to consider diplomatic, legal, and cyber defense measures. If the incentive is electoral victory, then media literacy campaigns, fact-checking, and platform accountability for political advertising become more salient. If it’s purely financial, then disrupting the advertising revenue streams of these profiteers can be a powerful deterrent. Without this foundational understanding, interventions risk being misdirected or ineffective.

The evolution of technology continues to reshape the landscape of actors and their incentives. The accessibility of sophisticated tools for content creation, amplification, and even deepfake generation means that the barrier to entry for aspiring manipulators is constantly lowering. A single individual with a laptop can now potentially reach millions, previously a capability reserved for well-funded organizations. This democratization of influence further complicates attribution and response, as the sheer volume of potential actors expands exponentially.

Moreover, the line between legitimate influence and manipulative disinformation can be incredibly fine, particularly in the realm of public relations and political campaigning. Advocacy groups, lobbyists, and even corporations engage in activities designed to shape public opinion and policy. While many operate within ethical boundaries, the tactics employed can sometimes overlap with those used by disinformation actors, such as the strategic seeding of narratives in sympathetic media, the use of astroturfing (creating fake grassroots movements), or the targeted dissemination of persuasive content. Distinguishing between legitimate, albeit biased, advocacy and malicious information operations requires careful scrutiny of intent, methodology, and transparency.

The shadowy world of "perception management" also plays a significant role, particularly for state actors and well-funded political campaigns. This involves not just spreading false information, but carefully curating and controlling the information environment to create a desired impression or narrative. It might involve suppressing inconvenient facts, promoting positive stories, or subtly guiding public discourse away from uncomfortable topics. The incentive here is to maintain a favorable image, control the narrative, and preemptively counter potential criticisms, often through a long-term, sustained effort rather than a single, high-impact disinformation blast.

Finally, the sheer human element of belief and bias is an incentive in itself. Individuals often gravitate towards information that confirms their existing worldviews, a phenomenon known as confirmation bias. This makes them more susceptible to narratives that align with their prejudices or hopes, regardless of factual accuracy. Disinformation actors skillfully tap into these inherent psychological vulnerabilities, crafting messages that resonate deeply with pre-existing beliefs, fears, and aspirations. The incentive for the recipient, in this case, is the validation of their own perspectives, making them willing, albeit often unwitting, amplifiers of manipulative content. This complex interplay of actors, their diverse incentives, and the inherent human susceptibility to biased information forms the bedrock of the news wars.

The digital landscape has fundamentally reshaped how political discourse unfolds, transforming social media platforms into the primary arenas where information, and disinformation, collide. This shift means that understanding elections and democratic trust in Europe requires a deep dive into the mechanics of these platforms: the algorithms that decide what users see, the advertising systems that micro-target voters, and the mechanisms that amplify certain narratives, often with unsettling consequences. These are not neutral spaces; they are engineered environments with profound implications for electoral integrity.

At the heart of every major social media platform lies a recommendation algorithm, a complex set of rules designed to keep users engaged. These algorithms prioritize content that generates high levels of interaction—likes, shares, comments—over content that might be more factual or nuanced. The unintended consequence is often the amplification of sensational, emotionally charged, or polarizing political material, simply because it tends to grab attention more effectively. This engagement-driven model can distort political narratives, giving disproportionate visibility to extreme viewpoints and conspiracy theories.

Studies in Europe have repeatedly shown how these algorithms can skew political discourse. For example, research into the 2025 German federal elections on TikTok and Instagram revealed that these platforms frequently displayed political content, with a notable prominence of extreme right-wing material. This wasn't always a reflection of user preference; even when simulated users (avatars) expressed interest in left-leaning politics, right-wing perspectives continued to dominate their feeds, suggesting that platform amplification can override individual choices. The algorithms, in essence, were pushing users towards more radical content.

This algorithmic bias isn't unique to Germany. A study across Finland, France, and Romania using avatars mimicking young adults (18-24) on TikTok, Instagram, and X (formerly Twitter) found that right-wing content constituted 58% of all politically classified posts, compared to just 26% for left-wing and 16% for centrist content. The consistent amplification of certain political leanings, regardless of initial user interest, raises serious questions about the fairness and balance of electoral debates within these digital spaces.

The implications extend beyond mere content visibility. When algorithms continually feed users similar content, it creates "filter bubbles" and "echo chambers." In these curated environments, individuals are primarily exposed to opinions that mirror their own, making it harder to encounter diverse perspectives or engage in meaningful political discourse. This can lead to increased political polarization, where different groups become entrenched in their own beliefs and view opposing sides as misinformed or malicious.

Beyond organic content amplification, political advertising on these platforms introduces another powerful, and often opaque, layer of influence. Social media advertising allows political actors to reach a broad audience at a relatively low cost and, crucially, to target specific voter groups with precision. This "microtargeting" uses data collected by the platforms to deliver highly tailored political messages to individuals based on their demographics, interests, and online behavior.

The use of microtargeting in political ads has raised significant concerns about democratic integrity. It enables campaigns to disseminate messages that might go unchecked by the broader public, as only specific, susceptible groups see the ad. This makes it easy to spread misinformation to targeted audiences with little accountability. The Cambridge Analytica scandal, which involved the improper access of millions of Facebook users' data for targeted political advertising during the 2016 US election and Brexit referendum, brought this issue to international attention.

In response to growing concerns, the European Union has moved to regulate political advertising. The EU's Transparency and Targeting of Political Advertising (TTPA) regulation aims to increase transparency by requiring platforms and advertisers to disclose who funds political ads, what elections or issues they concern, and to limit the use of personal data for targeting. This regulation, adopted in March 2024 and fully applicable by October 2025, sets strict rules on the use of microtargeting techniques that involve processing personal data.

The introduction of the TTPA has had a significant impact on platform behavior in Europe. Citing "unworkable requirements and legal uncertainties," Meta (owner of Facebook and Instagram) announced it would cease running all political, electoral, and social issue ads in EU countries starting in October 2025. Google's parent company, Alphabet, made a similar move in late 2024. While this decision aims to address regulatory burdens, it also means that online political speech in the EU will become even more algorithmically mediated, with political actors having less control over the reach of their paid content.

The move by major platforms to ban political ads in the EU, while intended to address regulatory concerns, may have unintended consequences. It shifts political communication further into the realm of "organic" content, where algorithmic amplification rules supreme. This could further disadvantage "normal" politicians who rely on paid reach for policy discussions, while potentially extending the advantage of those whose content naturally generates high engagement. The distinction between legitimate advocacy and manipulative content becomes even more blurred when everything operates under the same algorithmic logic.

The problem of amplification is particularly acute on platforms favored by younger demographics, such as TikTok. A study concerning Europe's political future observed that TikTok's algorithms push political content into users' feeds much faster than on platforms like Instagram. Evidence from the Romanian presidential election in December 2024, for instance, suggested that foreign actors mounted a coordinated TikTok campaign to garner support for a pro-Russian candidate, leading the European Commission to investigate whether TikTok had failed to mitigate systemic risks under the Digital Services Act (DSA). Similarly, in the Polish presidential election, the TikTok algorithm heavily favored right-wing content.

The DSA is a significant piece of EU legislation designed to hold large online platforms accountable for the content they host. It mandates that platforms identify and handle "illegal content" and introduces extensive requirements regarding content moderation practices. However, concerns remain about the effectiveness of these regulations in curbing algorithmic amplification of harmful content. MEPs, for example, have expressed alarm over algorithms pushing polarizing political messages and have demanded investigations into whether recommender systems undermine democracy and violate EU digital rules.

The challenges are multifaceted. Even when platforms provide transparency tools for political ads, they have historically been incomplete or underdeveloped, and self-regulatory codes often lack sanctioning mechanisms. The sheer volume of content, combined with the opaque nature of algorithms, makes it incredibly difficult to monitor and assess the full impact of these platforms on electoral outcomes. The rise of AI-generated content further complicates matters, as it can be used to create realistic images, videos, and audio that spread rapidly through algorithmic recommendation systems, amplifying polarization and undermining democratic discourse.

Indeed, Europol predicted in 2025 that 90% of online content could be synthetically generated using artificial intelligence, which would significantly accelerate the spread of sensationalism and misinformation, influencing public opinion based on engagement rather than accuracy. This means that while traditional forms of disinformation remain a threat, the increasing sophistication of AI tools allows for the creation of deepfakes and fabricated stories that can target political figures and be amplified through influencer networks, as seen in the "Storm-1516" operation during Germany's early 2025 parliamentary elections.

The dynamic interaction between algorithms, targeted advertising, and amplification fundamentally alters the information ecosystem during elections. Platforms, once seen as neutral conduits, are now recognized as powerful gatekeepers that shape civic discourse through their algorithmic mechanisms. For many young people, social media is their primary source of political news, meaning the content they encounter is optimized for engagement rather than democratic deliberation. This environment cultivates a feedback loop where extreme or emotionally charged ideas gain more traction, potentially leading to a more fragmented and polarized society.

The economic model of these platforms also plays a role. Advertising-driven business models often reward divisive and emotionally charged content over dialogue and compromise, viewing users not as citizens but as products whose data fuels revenue. This inherent incentive structure can inadvertently align with the goals of disinformation actors, regardless of their original intent. The platforms' capacity to influence behavior and steer public debate through opaque algorithmic mechanisms makes them central to any discussion about electoral integrity in the digital age.

Despite the challenges, efforts are underway to address these systemic issues. The DSA, for example, aims to ensure algorithmic transparency and stronger oversight to rebuild trust and improve the quality of online political debate. However, the implementation and enforcement of such regulations are complex, and platforms continue to navigate the boundaries between free expression and the spread of harmful content. The debate continues on how to ensure these powerful digital spaces support healthy democracy rather than contribute to its erosion.

The concern extends beyond overt disinformation. Algorithms can subtly influence perceptions by limiting users' exposure to diverse viewpoints, thereby strengthening existing biases. This constant reinforcement of beliefs, even without explicitly false content, can contribute to a breakdown of common ground in public debates and make societies more susceptible to manipulation. The very architecture of these platforms, therefore, becomes a battleground for the hearts and minds of European citizens.