A critical paradigm shift has taken place in cybersecurity.
Attackers no longer use AI merely as a productivity tool for creating phishing emails; they are now actively deploying adaptive, AI-powered malware in live operations. In this 60-minute briefing, we will cover the following topics:
- The Evolution of Adversarial AI:How Threat Actors Use “Just-in-Time” Code Generation, Autonomous Malware (e.g., PROMPTFLUX and PROMPTSTEAL), and Illegal AI Marketplaces on the Dark Web.
- Emerging Internal Vulnerabilities:The Hidden Risks Posed by “Shadow AI,” Model Extraction Attacks (Distillation), and Design Flaws in Agent-Based Systems (such as prompt injection and excessive decision-making authority).
- Real-world case studies:Current campaigns that use fake AI websites as a weapon and exploit legitimate AI platforms through deceptive “ClickFix” tactics.
- Securing the Future:Strategies for countering these threats through regular “AI red teaming” and ways to use AI as a force multiplier for one’s own defense (the “agent-based SOC”).
Speaker: - Prof. Dr. Igor Podebrad – Director, Office of the CISO, Google Cloud, and former Group CISO at Commerzbank
- Mike Hart – Google Cloud Security Manager for CEE/Alps
❗️ This event isopen exclusively to ITHM members.