Dark LLMs: When Your AI Traffic Is C2
Dark LLMs are writing per-host pwsh one-liners, self-rewriting droppers, and hiding in model APIs you approved. If you’re not policing AI egress, you’re not doing detection. 😬🤖
Dark LLMs are writing per-host pwsh one-liners, self-rewriting droppers, and hiding in model APIs you approved. If you’re not policing AI egress, you’re not doing detection. 😬🤖
UNC6485 is farming Triofox: Host: localhost → setup → mint admin → AV path = your script → SYSTEM → RMM + reverse RDP/443. Patch to 16.7.10368.56560 now. Copycats next. 🔥🛡️
Thailand pulled the plug. The grift brought generators + Starlink. Shift north→south (Shwe Kokko/Myawaddy; Tachileik/Mae Sai). Squeeze OTC cash-outs + first-funding friction, or watch it respawn.
VoidProxy: AitM Phishing-as-a-Service Quietly Bypasses MFA at Scale
If your bulk email or CRM gets popped, PoisonSeed rides your good reputation straight past filters and users’ instincts. Here’s the fast path to detect and blunt it—without boiling the ocean.
UNC3944, UNC6040, and UNC6395 are executing targeted campaigns against SaaS, cloud, and virtualization environments, leveraging vishing, OAuth abuse, and supply-chain compromise. Their TTPs require precise, telemetry-driven controls and detection.
Shamos, a new Atomic macOS Stealer (AMOS) variant attributed to COOKIE SPIDER, is targeting U.S. tech and education sectors via malvertising and fake support sites.