CyberzSOC

The purpose of this publication is to provide organisations with guidance on how to use AI systems securely. The paper summarises some important threats related to AI systems and prompts organisations to consider steps they can take to engage with AI while managing risk. It provides mitigations to assist both organisations that use self-hosted and third-party hosted AI This publication was developed by the Australian Signals Directorate’s Australian Cyber Security Centre (ASD’s ACSC) in collaboration with the following international partners: the Secretariat of Science, Technology and Innovation Policy, Cabinet Office The guidance within this publication is focused on using AI systems securely rather than developing secure AI systems. The authoring agencies encourage developers of AI systems to refer to the joint Guidelines for Secure AI System Development.