Framing the stakes

We reside in a moment in which technology increasingly shapes civic life, the integrity of democratic processes, and the role of civil society. Technology has shifted from tooling to foundational infrastructure for access to information, formulation of political opinions, and participation in civic life. It has transformed how democracy functions in an essential way. Thus, whether the outcome of this interconnection is positive or negative hinges on how we design, govern, and deploy digital systems and solutions. From online information ecosystems to secure civic infrastructure and equitable digital access, technology holds both promise and peril. Yet too often, technological development advances without sufficient grounding in human rights, public accountability, or democratic values. In countries like Bulgaria, democratic resilience faces challenges such as weak media pluralism and widespread disinformation that create fertile ground for polarization and mistrust. Recent evaluations highlight delays in implementing key EU digital regulations like the Digital Services Act (DSA) and the EU Artificial Intelligence Act (AI Act), alongside a lack of a national strategy against disinformation, weak criminalisation of online and technology-facilitated abuse, and ongoing pressures on independent journalism and civil society mobilisation.

Europeans consume social and political information through social media as the second most popular source – citizens both intentionally seek out content (66%) and unexpectedly encounter it during casual browsing (76%). In 2025, more than a third of respondents in the EU (36%) said they were ‘often’ or ‘very often’ exposed to disinformation and fake news over the past seven days. Exposure is rising, and confidence in identifying false information remains low. At the same time, social media has become the dominant news source for younger generations, concentrating informational power within opaque, algorithm-driven environments.

In Bulgaria, these trends are particularly acute. The country consistently ranks among the lowest in the EU for media freedom and trust in institutions. According to the Reuters Institute for the Study of Journalism Digital News Report 2025, Bulgarians get their news from online sources like news websites, social networks, news podcasts and AI chatbots (74%), as well as social media (57%). However, only 26% of Bulgarians trust the news overall, one of the lowest levels in Europe. Moreover, 31.4% of adult Bulgarians have used Generative AI, but 42.2% are worried about the technology.

Against this backdrop, this article is framed around a central question:
How can digital systems be reshaped to strengthen, rather than destabilise, democratic life? We answer it by surfacing key insights from leading voices in responsible tech and governance like David Ryan Polgar (All Tech is Human), Niki Iliadis (The Future Society), and Rumman Chowdhury (Humane Intelligence).

Technology as both enabler and destabiliser of democracy

David Ryan Polgar’s work consistently stems from a simple but powerful premise: “All tech is human. The future of tech is intertwined with the future of the human condition and the future of democracy.” The current tech ecosystem, however, has been structured around engagement-driven, bottom-line informed incentives, often at the expense of truth, trust, and democratic integrity. As Polgar and colleagues write, “one of the well-known risks of generative AI is the proliferation of deepfakes. There are alarming implications of synthetic media for everything from the integrity of elections to celebrity nudes. […] Beyond individual pieces of fabricated content, generative AI is being used to create deepfake websites, thereby making it easier to create and distribute additional false content at scale.” This creates a systemic challenge for human rights, democracy, and public trust.

Recent trends confirm the scale of this challenge:

• Over one-third of Europeans report frequent exposure to disinformation.
• Social media accounts for the overwhelming majority of its spread.
• GenAI use for personal, information-seeking uses is rising.

In Bulgaria, the destabilising effects of this dynamic are particularly visible. The country ranks among the lowest in the EU in the Reporters Without Borders World Press Freedom Index, reflecting structural vulnerabilities in the media environment. Disinformation ecosystems in Bulgaria are often:

• Highly politicised, with links between media ownership and political actors
• Amplified through Facebook-dominated information channels, where a significant share of the population consumes news
• Influenced by foreign narratives, particularly pro-Kremlin messaging in the context of geopolitical tensions.

Thus, technologies and online platforms that have the capacity to enable civic mobilisation, knowledge exchange, and the formation of interpersonal connections, also amplify manipulation, spread harm, and serve as a malicious tool. If the problem is structural, the response must be systemic. So, what could a pathway to healthier, more democratic digital ecosystem look like? Polgar argues that “technology is intertwined with society, with human behaviour,” hence the involvement of diverse perspectives – from technologists to ethicists and psychologists – is necessary (All Tech Is Human). Amplifying the voice of civil society actors that contribute evidence-based insight into the health of the digital ecosystem, and actionable expertise on the design and development of human-centric technologies is a critical step to, first, understanding and, consequently, tackling such systemic challenges.

From pro-innovation to a “whole-of-society” governance model

Post-2023 discourse on tech governance has overwhelmingly centred on pitching two “opposite” views against each other – pro-innovation approaches versus pro-regulation positions. This unhelpful dichotomy has led to polarisation of public opinion on how online spaces, technologies like AI, and digital actors should be framed for the benefit of society. Cross-sectoral initiatives like the Global Call for AI Red Lines, which has been endorsed by over 90 organizations and 1,500 individuals, illustrate the necessity to transcend binary conversations and focus on a “whole-of-society” approach. As Niki Iliadis, the Director of Global AI Governance at The Future Society, described it, “red lines are not supposed to be about panic, they’re more about prevention […] and they’re definitely not about slowing down innovation.” “Red lines are about protecting society,” Iliadis says, “The alternative would be to wait for more large-scale harm to actually manifest, and that’s not responsible. The world has drawn red lines before […] The call is to get the international community to do it again for AI. The window is narrowing, but this is an opportunity for governments to come together to find that lowest common denominator of what isn’t working, and then to understand that some risks from AI are just too great to take, and that we need some lines that must never be crossed.”

Despite regulatory advances such as the EU’s Digital Services Act and the EU AI Act, implementation gaps remain. Evidence suggests that platform and tech provider responses to systemic risks, particularly during elections, moments of socio-political turbulence, and in the context of sensitive topics, are uneven and insufficient. For example:

• Recent whistleblower reports show that social media giants like Meta and TikTok made decisions which allowed more harmful content, including misogyny, violence, sexual blackmail, conspiracy theories and terrorism, on people’s feeds as part of the design choices behind engagement algorithms (BBC).
• AI chatbot providers like OpenAI and Google are facing an increasing number of court cases in relation to young people’s mental health, including wrongful death allegations. As of late 2025, OpenAI appeared in at least nine different lawsuits “alleging wrongful death, assisted suicide, involuntary manslaughter, and a variety of product liability, consumer protection, and negligence claims.”
• Global Witness’s global study among land and environmental defenders from 2025 “found that online abuse is very common among defenders […], and frequently translates into offline harm, including harassment, violence and arrests.”

Worldwide, the harms of the poorly governed digital ecosystem affect not only activists and defenders, but regular users and vulnerable audience – the issue is not isolated. In Bulgaria, the need for coordinated governance is further underscored by institutional fragmentation and policy delays. Key challenges include:

• Delayed and uneven implementation of EU digital regulations, including aspects of the Digital Services Act and the AI Act
• Absence of a comprehensive national strategy on disinformation and laws against the full spectrum of online and technology-facilitated abuse
• Limited coordination between government, regulators, and civil society actors

At the same time, civil society initiatives, such as independent fact-checking organisations and media literacy programs, have emerged as critical actors, often filling gaps left by state institutions. Bulgaria has an acute need for ecosystem-wide alliances and spaces for evidence-creators to voice their positions and be heard by political decision-makers.

Embedding democratic principles into technological design

If governance is to catch up with technology, it cannot remain purely reactive. Apart from the unhelpful pitching of innovation against governance discussed earlier in this article, Rumman Chowdhury, CEO and co-founder of Humane Intelligence, coined the phrase “moral outsourcing” to describe the application of “sentience and choice to AI, allowing technologists to effectively reallocate responsibility for the products they build onto the products themselves – technical advancement becomes predestined growth, and bias becomes intractable.” Beneath the surface narratives like this reflect foundational design choices, i.e., the conscious preference to remove tech creators from having accountability for the outcomes of the use (or misuse) of their technologies. But when “generative AI content is already shaping and skewing our perceptions of the world, and how that’s increasingly related to election or political content” (Greenpeace), we need to pay increasing attention to design-stage accountability.

The urgency of this shift is reinforced by emerging evidence:

• 89% of journalists believe AI will significantly increase disinformation risks, particularly through deepfakes and synthetic media. 
• Disinformation is increasingly coordinated, networked, and strategically deployed, often linked to geopolitical actors. 
• AI-driven disinformation (e.g. deepfakes) is expected to significantly increase manipulation risks and the potential to entrench systemic digital abuse.
• Digital systems are becoming more complex, less transparent, and harder to regulate.

In Bulgaria, the risks associated with emerging technologies and online spaces are compounded by low levels of digital and AI literacy and institutional trust. According to the European Commission Digital Economy and Society Index (DESI):

• Bulgaria ranks among the lowest in the EU in digital skills, with a significant share of the population lacking basic competencies.
• This creates vulnerabilities to AI-generated misinformation and manipulation, particularly as synthetic media becomes more widespread.

These conditions make design-stage safeguards, such as transparency, human intervention, and accountability mechanisms, even more critical in fragile contexts like Bulgaria. The definition of these design choices in line with democratic principles cannot be the job of the tech industry alone. Civil society – as a key part of the digital ecosystem – is a critical actor that can effectively surface localised, data-backed insights, formulate meaningful practical steps towards healthier and safer tech and online spaces, and provide democratic oversight, ongoing monitoring, and evaluation of efforts. But civil society, especially in Bulgaria, requires enabling conditions: commitment to interdisciplinary engagement on part of policymakers, cross-sectoral cooperation across the developer -> deployer -> user spectrum, and structural reforms that redistribute power within the digital ecosystem (The Future Society, Project Liberty).

Conclusion: from disruption to democratic design

What emerges across these perspectives is not simply a critique of current digital systems, but a roadmap for transformation. Democratic resilience in the digital age depends less on regulation alone and more on redesigning the digital ecosystem through multi-stakeholder, civil society-driven approaches.

The challenge is simultaneously to mitigate risks such as disinformation or algorithmic bias, while also redefining the relationship between technology and democracy. This requires moving beyond reactive governance toward intentional design, collective responsibility, and institutional innovation.

As Polgar, Iliadis, and Chowdhury each demonstrate in different ways, the future of digital systems is neither fixed nor inevitable. It is shaped by choices: about incentives, governance structures, and who gets to participate in decision-making.

In a European context marked by rising disinformation, uneven regulatory implementation, and fragile trust, the stakes are particularly high. But so too is the opportunity: to build digital systems that do not merely coexist with democracy but actively sustain and strengthen it.

This is a vision that The Counterintuitive Institute has pledged to support in our Strategy 2030 by creating spaces for cross-sectoral discussion and collaboration like the Counterintuitive Summit, expanding networks and collective action beyond “traditional” players, and amplifying progressive, impactful voices through open, public resources. The democratic stakes of our digital futures create a strong incentive. To drive positive disruption.