
Attackers use CyberStrikeAI to direct intrusions at schools
Kaspersky says a human operator used an LLM on the C2 server to pick tools while Go beacons ran commands on compromised hosts.
12 stories, newest first · one card per story

Kaspersky says a human operator used an LLM on the C2 server to pick tools while Go beacons ran commands on compromised hosts.

Operators can dump credentials, log keystrokes, call OpenAI from the host, and deploy screen-lock ransomware.
A leetspeak insult password was a real domain admin credential; the outdated domain-joined machine, not a weak password, was the entry point.

Attackers abused a Viva Aerobus SQL Server to run commands and steal files, then exposed tools and data online.

ThreatMon exposed an unauthenticated attacker staging server revealing MSSQL intrusion tooling and stolen data linked to a Viva Aerobus environment.

A local uncensored Qwen model revised an LSASS dumper until two lab EDRs no longer detected it.

An uncensored local Qwen derivative generated an LSASS dumper that evaded two EDRs in lab tests.

Microsoft says affiliate Storm-2570 reused one playbook across Qilin, DragonForce, Anubis, and BERT.

Trellix details how attackers dump the NTDS.dit Active Directory database via shadow copies to harvest NTLM hashes and forge persistent Golden Tickets.

Attackers abuse Windows VSS shadow copies to delete recovery points before ransomware and to extract NTDS.dit/SAM credentials, demanding context-aware detection.
Huntress explains how attackers steal credentials through phishing, AitM, infostealers, and dumping, then use them for lateral movement, BEC, and ransomware.

AI-agent campaign exploited PaperCut CVE-2026-81578 and CVE-2026-82078, compromising 440 servers at 395 organizations and reaching Domain Admin in 12.
Today 1.6K readers · 2.4K views · 775 reads · all time 189K views