Solutions stage
Panel
In person & online
LeadersGoldDiscovery

AI and democracy: Threats, safeguards, and the path forward

  • Date
    7 July 2026
    Timeframe
    09:30 - 10:00 CEST
    Duration
    30 minutes
      Democratic institutions exist to translate the will of citizens into accountable decisions. Courts, parliaments, regulatory agencies, and public media all rest on one condition: that citizens can verify who made a decision, on what basis, and through what process.

      That verification chain is now under structural pressure. Public servants are increasingly relying on AI systems whose values were set unilaterally by private developers, often in another jurisdiction and with no democratic mandate. Recent documented incidents have shown that a single unauthorized change to an AI system’s instruction layer can silently alter the information received by millions of users at once. This pressure is also a matter of European sovereignty in AI: when the models shaping public decisions are built, governed, and updated beyond the reach of the institutions they affect, governance is compromised before any individual decision is made.

      This panel will discuss a practical response: internationally interoperable AI certification standards that rate deployed models on their adherence to human rights and democratic integrity, designed to give every nation a common basis for democratic accountability.

      Safeguarding democratic institutions in the age of AI is a shared responsibility. This session brings AI for Good participants into that conversation, which depends on cooperation across governments, researchers, and industry, and on a shared commitment to the safe and accountable use of AI.

      Co-organized by the Center for International Governance Innovation (CIGI).

      Are you sure you want to remove this speaker?