Can democracy speak under AI-driven visibility? Toward an affective–algorithmic reworking of the spiral of silence

Abstract

This article offers a critical theoretical reassessment of the conditions shaping political expression and its boundaries within algorithmically mediated publics, examining how artificial intelligence–driven infrastructures of visibility, intertwined with affective dynamics, alter the conditions under which opinions acquire public presence and come to be perceived as legitimate to express. The argument is illustrated through a case-based analysis of highly visible user interactions surrounding posts by the Portuguese populist leader André Ventura across Facebook, TikTok and X, which serves to explore how different socio-technical environments give rise to distinct climates of opinion. Revisiting the Spiral of Silence theory from the perspective of contemporary platform environments, the article argues that silencing is no longer adequately explained primarily through interpersonal fear of isolation, but must instead be understood as emerging from the interplay between algorithmically structured visibility and affectively charged interaction. On this basis, the article reconceptualises the Spiral of Silence as an affective–algorithmic process and contributes to debates in communication theory and digital political communication, while asking whether, and under what conditions, democracy can still be said to speak when publicness itself is algorithmically organised.