LATIDIA · Ciberseguridad

PIA-Bench: Hacia la evaluación automatizada del impacto en la privacidad con modelos de lenguaje extensos

arXiv:2609.12571v1 Anuncio Tipo: nuevo Resumen: La evaluación del impacto en la privacidad (PIA) es un instrumento fundamental para que las instituciones identifiquen de forma proactiva los riesgos de privacidad y desarrollen estrategias de mitigación antes de la implementación del sistema.

La noticia

arXiv:2609.12571v1 Announce Type: new Abstract: Privacy impact assessment (PIA) is a critical instrument for institutions to proactively identify privacy risks and develop mitigation strategies before system deployment. While mandated across regulatory and institutional contexts, executing PIA requires extensive privacy and technical expertise, posing a particular challenge for teams without access to such resources. Prior work shows the potential of leveraging large language models (LLMs) to assist practitioners' privacy decisions, but little is known about how accurately and reliably LLMs can automate PIA. To this end, we develop PIA-Bench, the first open benchmark for evaluating LLMs on real-world PIAs. We first audited 499 expert-authored PIAs published by US federal agencies and