Name
A framework for risk management in systems with AI components
Authors
Jose Manuel Camacho and David Arroyo
Abstract
The upcoming release of the NIST Artificial Intelligence risk management framework and the
European Union AI Act raises new and intriguing questions about the risk analysis of AI-based
systems. To address these issues, this study presents a comprehensive cybersecurity risk
management framework for information systems incorporating AI components. The proposed
framework provides a scheme to simulate potential attacks on these systems and assess their
impacts. Additionally, we present a cyber mitigation selection approach for risk management
purposes. We conclude with a case study related to protecting Autonomous Driving Systems.