MITRE ATT&CK v20 landed October 13, and for the first time AI abuse is mapped as real techniques with real IDs. Tova Dvorin and Adrian Culley break down T1698 and T1699 — indirect and direct prompt injection — plus AI-enabled exfiltration and the new technique for impairing the LLM assistants your SOC now leans on. Then two new campaigns: an AI-generated 49-module toolkit used against Mexico’s government, and 30+ Minnesota water utilities locked out through internet-facing PLCs. A taxonomy is a map, not a test.
Podden och tillhörande omslagsbild på den här sidan tillhör
SafeBreach. Innehållet i podden är skapat av SafeBreach och inte av,
eller tillsammans med, Poddtoppen.