Alena Birrer

When people hear the word “deepfake” in the context of politics, they most likely imagine something sinister. They might picture a politician's face convincingly placed onto someone else's body, a beloved representative framed for corruption, or even the fabrication of a scandalous tape aimed at sullying the reputation of an opponent. While malicious content certainly exists and poses genuine risks to democratic discourse, our research on political deepfakes tells a more nuanced story.

Social-media platforms often limit the visibility of problematic content and accounts instead of removing them entirely. While visibility reduction seeks to balance free speech and safety concerns, it has fueled criticism over platform power, potential bias, and shadow banning. This article defines visibility reduction, categorizes it into four main types and examines its application using data from the newly established Digital Services Act’s Transparency Database. The analysis highlights tha ...