Publication of the CLUSIT AI Security Report

In March 2026 the CLUSIT community published the AI Security report, a document devoted to AI security risks, emerging attack surfaces, and the controls needed by organisations that are bringing models and AI agents into real operational settings.

Federico Cerutti contributed to the report within the CLUSIT Community for Security, bringing into the document the Brescia node perspective on reliability, risk assessment, and the interplay between technical and organisational controls.

The report turns themes such as reliability, risk assessment, and organisational controls into a practical reference for people responsible for adopting AI systems in operational and regulated settings.