Aligning Trustworthy AI with Democracy: A Dual Taxonomy of Opportunities and Risks

AI poses dual threats and opportunities to democracy: The paper identifies specific risks including threats to voter autonomy, fairness in political processes, and erosion of public trust, while also highlighting AI's potential to improve government transparency, increase citizen participation, and enable evidence-based policymaking.

Transparency and societal well-being are critical across all AI applications: The research emphasizes that these two factors cut across every category of democratic risk, suggesting they should be prioritized as foundational principles in any responsible AI governance framework.

The EU's Trustworthy AI framework provides actionable mitigation strategies: Rather than simply identifying problems, the paper grounds each identified risk in concrete mitigation approaches based on European regulatory standards, offering policymakers and institutions practical tools for oversight and system design.

A structured taxonomy helps different stakeholders align on democratic values: By creating parallel frameworks for both risks and opportunities, the paper provides scholars, policymakers, and technologists with a common language and evaluation criteria to assess whether AI systems support or undermine core democratic principles like autonomy and fairness.

Responsible AI requires integration of democratic theory with technical governance: The paper bridges the gap between abstract democratic ideals and practical system design, suggesting that trustworthy AI cannot be achieved through technical solutions alone but requires deliberate alignment with democratic values at every stage of development and deployment.