Il paradosso dell'intimità artificiale: come l'IA frammenta e riassembla i legami sociali

The advent of artificial intelligence (AI) has introduced an epochal transformation in the structuring of social bonds, redefining not only the modes of interaction but the very ontology of human connection. This article critically explores the paradox of artificial intimacy (Turkle, 2015): on one hand, AI promises more efficient, personalized, and inclusive connections; on the other, it fragments organic bonds, replacing them with algorithmic dynamics that reproduce power asymmetries, surveillance, and alienation. Through a qualitative analysis of public discourses (CEO statements, policy documents) and emblematic case studies (dating platforms, social robots in caregiving, virtual influencers), the study examines three key dimensions: the commodification of intimacy (how algorithms transform emotions and relationships into extractable data, fueling surveillance capitalism (Zuboff, 2019)); the uman-machine hybridization (the emergence of non-human social actors (Latour, 2005) that reshape hierarchies, roles), and relational expectations and the crisis of agency, the delegation of intimate choices to opaque systems, resulting in the atrophy of empathetic competencies (Turkle, 2011). The findings reveal that AI is not merely a technical tool but an *emerging social institution* that structures norms, sanctions deviations, and naturalizes logics of control. To mitigate the risks of dehumanization, the article concludes by proposing building on existing theoretical contributions (Feenberg, 2002; Suchman, 2002; Benanti, 2022) a model of participatory sociotechnical ethics, in which democratic regulation, critical education, and inclusive design aim to reconcile innovation with the authenticity of social bonds.