Disinformation as a Security Signal: Rethinking Assumptions for AI-Driven Democratic Resilience

Much of the literature (e.g., Romănishyn et al., 2025; Allegri, 2024) assumes disinformation is always a threat to democracy, justifying securitization and regulatory interventions. But what if, in some contexts, exposure to disinformation paradoxically strengthens democratic processes—by mobilizing civil society, prompting media innovation, or catalyzing critical thinking? This research would use AI to detect patterns where spikes in disinformation correlate with positive democratic outcomes (e.g., voter turnout, fact-checking activity, policy reforms). By surfacing these "unexpected outcomes," the project challenges core assumptions about disinformation's role and explores how AI can distinguish between destabilizing and constructive effects. This could inform a more sophisticated, context-sensitive approach to AI interventions—one that supports, rather than stifles, organic democratic resilience.

References:

1. AI-driven disinformation: policy recommendations for democratic resilience. Alexander Romănishyn, Olena Malytska, Vitaliy Goncharuk (2025). Frontiers Artif. Intell..
2. The Impact of Disinformation on the Functioning of the Rule of Law and Democratic Processes in the Eu. M. Allegri (2024). Interdisciplinary journal of research and development.