About the ACSC
The Australian Cyber Security Centre (ACSC), is the Australian Government's lead agency for cyber security. The ACSC is part of the Australian Signals Directorate and is based at the Australian Security Intelligence Organisation headquarters. The ACSC leads the Australian Government's efforts to improve cyber security. It brings together cyber security capabilities from across the Australian Government to improve the cyber resilience of the Australian community and support the economic and social prosperity of Australia in the digital age.
The ACSC's goal is to facilitate information sharing, raise awareness of cyber threats and to minimise security risks surrounding the adoption of new technologies and services.This organisation provides cyber security advice and services to government, critical infrastructure, industry and the Australian public.
Why AI Security is Critical for ACSC
Cybercriminals are adapting to capitalize on new opportunities, such as artificial intelligence, which reduces the level of sophistication needed for cybercriminals to operate. Restricting privileged access to the system mitigates several threats, including data poisoning and model theft. The guidance provides insights into choosing secure and verifiable technologies. For ACSC, AI security represents both an emerging threat vector that cybercriminals are exploiting and a critical capability that organizations must deploy securely to maintain Australia's position as a secure place to connect online.
Mileva delivered an AI-security workshop to the Cyber Uplift Branch that catered to both technical and business professionals expanding their cyber expertise into the AI domain. The workshop delved into advanced AI security concepts, threat modeling specifically for AI systems, techniques for detecting and mitigating adversarial machine learning attacks, and considerations for securely integrating AI within existing cybersecurity frameworks. Drawing on Harriet Farlow's extensive background at the Australian Signals Directorate (ASD) and the technical expertise of Miranda R, a Mileva team member and experienced AI security researcher and penetration tester, the session provided attendees with practical, actionable insights tailored explicitly to national security contexts. The interactive format and expert-driven discussions ensured that participants left equipped with the knowledge to effectively address evolving AI-related threats within their operational environments.
