Lessons from Penetration Tests on Large-Scale Agent Systems

As AI systems gain increasing autonomy and execution capability, the number of discovered security vulnerabilities continues to rise. However, many of these vulnerabilities are not fundamentally novel, but instead reflect recurring classes of weaknesses long observed in prior computing systems. Execution-capable AI agents are effectively unbounded, self-modifying programs that interact extensively with multiple layers of the computing stack. This broad interaction surface imposes a significant security burden on developers, who must reason about and secure complex cross-layer behaviors. Prior research has primarily focused on vulnerabilities in open-source agents and agent frameworks. In contrast, it remains unclear whether proprietary agent systems -- developed under stricter coding standards and formal review processes -- exhibit similar security weaknesses. In this paper, we present findings from two penetration tests conducted in 2025 against proprietary agent products and evaluate whether the security posture of AI agents has improved since these assessments.

Evidential Cyber Threat Hunting

A formal cyber reasoning framework for automating the threat hunting process is described. The new cyber reasoning methodology introduces an operational semantics that operates over three subspaces -- knowledge, hypothesis, and action -- to enable human-machine co-creation of threat hypotheses and protective recommendations. An implementation of this framework shows that the approach is practical and can be used to generalize evidence-based multi-criteria threat investigations.