<p>This article offers a critical theoretical reassessment of how political expression, dissent, and silence are structured within algorithmically mediated publics.Moving beyond the assumption of digital platforms as neutral arenas of participation, the article examines how AI-driven visibility infrastructures and platform-specific interactional cultures reshape the conditions under which opinions become visible or legitimate to express, as well as the risks associated with their articulation. Drawing on an illustrative, case-based analysis of highly visible user interactions on Facebook, TikTok, and X surrounding posts by the Portuguese populist leader André Ventura, the article mobilises empirical material to illuminate how distinct sociotechnical environments generate differentiated opinion climates. Engaging critically with Spiral of Silence theory, the analysis suggests that contemporary silencing dynamics may be driven less by interpersonal fear of isolation than by the interplay between algorithmic filtering and emotionally charged interactional environments. By reconceptualising the Spiral of Silence as an affective–algorithmic process, the article contributes to debates in communication theory and digital political communication, while also reflecting on the normative implications of algorithmic visibility for both publicness and democratic participation, particularly in relation to platform governance.</p>