Artificial intelligence and machine learning have rapidly become central tools in modern cybersecurity – not as replacements for human expertise, but as powerful amplifiers of analysis, decision-making and defensive capability. In this keynote, Adel Bibi, Senior Researcher in Machine Learning at the University of Oxford, explores how advanced ML models are used to detect anomalies, behavioural patterns and threats that would otherwise remain unseen. Drawing on current research, the session explains how AI can identify previously unknown attacks, reduce response times and support security teams in increasingly complex and fast-moving threat environments. The focus is not only on what AI can do, but also on its limitations, and how organisations can apply these technologies responsibly and effectively in practice. AI is not a silver bullet, but when applied correctly it becomes a powerful tool to strengthen human judgement and build faster, more adaptive cyber defences.

 Key takeaways:
 - How machine learning is used to detect novel and unknown cyber threats
 - The difference between traditional rule-based systems and AI-driven approaches
 - How AI is integrated into SOC operations and incident response as decision support
 - The evolving “human + machine” model in cyber defence
 - Common pitfalls in AI implementation – and how organisations can avoid them