Workshop 3 – CISO Track – AI and Cybersecurity
Dr. Sharif Abuadbba, Senior Research Scientist - Cybersecurity, CSIRO's Data61
Recently, there has been an increase in the application of Artificial intelligence (AI) technologies to support cyber-attacks, such as cyber intrusion, data theft and data compromise.
Future AI-driven cyber-attacks are expected to be more precise, targeted and tailored by training data to activate only in the presence of specific information or other cues. They will also be adaptable to a much greater extent than traditional rule-based phishing and malware.
Traditional approaches of hardening defences are unlikely to yield success against these advanced threats on their own, as this strategy relies heavily on complete coverage of all system loopholes that attackers might exploit, at all times.
Malicious learning processes will also be harder to detect, by virtue of the sophisticated detonation mechanisms and ability to mimic benign processes and behaviours.
Employing AI for Cybersecurity to solve these challenges can be a double-edged sword as it can be used for both cyber defence and cyber-attacks. Besides, AI itself is not a silver bullet and can be exploited.
In this session, I’ll focus on two aspects:
- (a) presenting a few real-world examples of developing novel AI based algorithms for cyber defence,
- (b) exploring the security of deployed AI algorithms and developing corresponding countermeasures.