Another keynote at the IoTSF conference took the concept of AI in IoT security further, talking about autonomous cyber defence; Anna Knack, a researcher at the Alan Turing Institute’s Centre for Emerging Technology and Security (CETaS), outlined the research in this area. She provided some of the areas covered in their report, “Autonomous Cyber Defence – A roadmap from lab to ops,” produced jointly by Georgetown University’s Center for Security and Emerging Technology (CSET) and The Alan Turing Institute’s CETaS.
It looks at current state-of-the-art in autonomous cyber defence and its future potential, identifies barriers to progress and recommends specific action that can be taken to overcome those barriers. The findings and discussion are of relevance to cybersecurity practitioners, policymakers and researchers involved in developing autonomous cyber defence capabilities.
The study is anchored on the potential for reinforcement learning (RL)-based AI agents to provide the autonomous capabilities required to fulfil some or all of the autonomous cyber defence concept. While the breadth of promising and relevant modelling approaches, techniques and technologies that relate to autonomous cyber defence is large, the researchers’ focus on RL is guided by the increased efforts in applied RL for cyber defence and what they said is the promising results RL has achieved in other problem domains.
- Nitin Dahad
