Inside AI-Driven Intrusions: What GTG-1002 Means for Defenders
Join XBOW’s Head of Security Nico Waisman and Head of AI Albert Ziegler for a deep breakdown of Anthropic’s GTG-1002 campaign, the first publicly documented case of an AI agent completing most of an intrusion chain.
In this recorded webinar, you’ll learn what actually happened, where models excel and fail in real offensive workflows, and how XBOW safely mirrors this depth to help teams assess their exposure in days, not weeks.
What you’ll learn:
What GTG-1002 actually demonstrated: how an AI agent chained recon, exploitation, and lateral logic across multiple stages.
Where LLMs are strong, and where they collapse without structure: planning drift, non-determinism, state loss, and hallucinated “success.”
How XBOW replicates AI offense safely: governed architecture, validators, and a practical playbook defenders can use today.
Speakers
CISO | XBOW @ XBOW
Head of AI | XBOW @ XBOW
Related Resources
How Moderna is Rethinking Security Testing in the Age of Autonomous and AI-Driven Attacks
Nico Waisman
Autonomous Offensive Security at Scale: Modern Penetration Testing for Enterprises
Loi Liang Yang
