LATIDIA · Ciberseguridad

La privacidad en la IA personalizada es una propiedad del sistema, no solo una propiedad modelo

arXiv:2609.38289v1 Tipo de anuncio: nuevo Resumen: En aplicaciones de IA personalizadas, como asistentes conversacionales y sistemas de recomendación, los usuarios no interactúan con modelos de forma aislada, sino con sistemas más amplios que acceden

La noticia

arXiv:2609.38289v1 Announce Type: new Abstract: In personalized AI applications, such as conversational assistants and recommender systems, users interact not with models in isolation but with broader systems that access, infer, and reuse user information across components and over time. While such use of user information is integral to personalization, it also raises important privacy questions. In this paper, we argue that individual model- or component-level analyses may not capture all privacy risks arising in such systems, motivating a system-level perspective on privacy. We distinguish and analyze four interconnected privacy-risk channels in personalized AI, and subsequently propose four requirements for system-level privacy evaluation, covering interaction trajectories, internal information flows, indirect leakage,