Episode 208: October 26, 2026
This episode covers a sharp shift in the threat landscape — threat actors are now using agentic AI operationally during intrusions, not just to write code but to run live attacks. We also dig into how AI agents can be tricked into executing malicious commands through their legitimate permissions, and why the LiteLLM supply chain compromise is a wake-up call for developers moving fast in the AI tooling space. Stories covered: - PHP2Shell Exploitation Hits Government Entity Amid Global Scanning and Signs of AI-Assisted Hacking (The Hacker News) -…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.