Internet platforms must be held accountable for their actions

Summary

This commentary argues that internet platforms bear substantial responsibility for the global retreat of democracy and must be held accountable for the role they play in enabling authoritarian trends. Drawing on warnings from independent research institutions and on a catalogue of more than 2,300 actions by the second Trump administration that echo authoritarian regimes, Lewandowsky positions platform dynamics as one identifiable factor among the complex, multi-causal drivers of democratic backsliding. The piece is a normative call: accountability for platforms is framed not as a regulatory nicety but as a defence of democratic institutions.

Key Contributions

• Frames platform accountability as a central pillar of democratic defence rather than a peripheral regulatory concern.
• Synthesizes warnings from multiple independent research bodies documenting worldwide democratic backsliding.
• Connects contemporary U.S. political developments — concretely indexed by 2,300+ catalogued administration actions — to broader global authoritarian patterns.

Methods

Opinion/commentary piece. The argument rests on (a) prior historical analysis co-authored by Lewandowsky on structural drivers of authoritarianism, (b) reports from independent democracy-monitoring institutions, and (c) a third-party catalogue of administration actions resembling authoritarian playbooks.

Findings

• Multiple independent research institutions converge on a diagnosis of global democratic retreat.
• More than 2,300 actions by the second Trump administration have been documented as resembling authoritarian regime behaviour.
• Structural factors — including platform dynamics — are implicated in driving these trends, though causation is multifaceted.

Connections

This commentary sits in dialogue with research arguing that platform governance is now inseparable from democratic resilience, most directly with work on the structural harms of platforms and the case for independent oversight such as Bak-Coleman2025-pm and Bak-Coleman2026-mk. It also connects to scholarship on data access and accountability infrastructures needed to study platform effects — see Rieder2025-ju, Rieder2026-pp, and Ohme2026-nv — and to broader debates about platforms’ political role explored in Tornberg2025-ir and Munger2025-cz.