AI and Cybersecurity Report
A closed roundtable held on 24 June 2026 with more than 20 participants examines, through cases from the field, the fronts AI has opened in cybersecurity: from asymmetric warfare to shadow AI and from model governance to critical infrastructure security. The session's core diagnosis is that AI has magnified the asymmetry between attacker and defender to an enormous degree.
What is in the report?
Asymmetric Warfare and Autonomous Threats
While organizations spend months on tendering, testing and budgeting to buy defensive products, attackers can rent autonomous vulnerability-scanning tools on the dark web within seconds; SOC teams find it increasingly hard to keep up by human eye alone.
Model Security and Shadow AI
Employees uploading company data and code to public language models is the hardest data leak to detect; network-level bans largely fail once staff use a personal phone to ask AI about the data on their screen.
Decision Security and Governance
Reducing AI governance to KVKK compliance is a strategic mistake, and it is unclear who answers for an AI assistant that makes a mistake. Participants foresee cybersecurity evolving toward “Decision Security”, in which algorithms are audited.
Social Engineering and Critical Infrastructure
Generative AI has perfected phishing and the impersonation of executives through deepfakes; AI automation in energy grids calls for a “Critical Infrastructure AI Security” doctrine that reaches beyond standard policies.