SEC5xx — CYBER DEFENSE CURRICULUM
SEC5xx — Detecting and Responding to AI-Generated Adversary Content
WHERE SANS' CATALOG STOPS
| Course | What it covers |
|---|---|
| SEC450 | Analyst USES AI in the SOC |
| SEC598 | Team AUTOMATES with AI |
| SEC535 | Red team ATTACKS with AI |
| This course | Detection + response for adversary content GENERATED by AI |
WHAT 'AI-GENERATED ADVERSARY CONTENT' MEANS
Arup, Feb 2024, ~$25.6M
Real-time voice and video clones in CEO-fraud calls and Zoom/Teams meetings.
FraudGPT/WormGPT successors, 2024-2025
Fluent, locale-correct phishing at scale, A/B-tested by the operator's own LLM.
ESET PromptLock, Aug 2025
First documented ransomware calling an LLM at runtime to generate payloads per victim.
EchoLeak CVE-2025-32711, June 2025
Zero-click M365 Copilot data exfil via crafted email parsed by the assistant.
UK NCSC/CISA guidance, 2025
End-to-end agent operations by financially motivated and state actors.
LiteLLM/Mercor breach, Mar 2026
Compromised PyPI packages exfiltrating cloud credentials and ML pipeline secrets — Mercor breach scaled to ~4TB.
5-DAY COURSE STRUCTURE
Deployment decision, embeddings, RAG for detection engineering, plus the first threat class.
Arup case study, synthetic audio/video detection, out-of-band verification as detection.
LLM-authorship signals, polymorphic malware, the EchoLeak class, guardrail stack as detection telemetry.
Detect adversary agent telemetry, harden your own agents, audit ML artifact provenance.
8-hour immersive IR against PROMETHEUS-7 attacking Verdancy Health.
ON THE SEC450 GRADUATE PATHWAY
PREREQUISITE
SEC450
Blue Team Fundamentals & SOC
THIS COURSE
SEC5xx
AI-Generated Adversary Content
This course assumes SEC450 as prerequisite. It does not re-teach the RAG, agent, or Ollama labs that SEC450's 2025 refresh already covers. Day 1 begins where SEC450 ends.
The curriculum positioning is intentionally aligned with the SEC450 graduate pathway — designed in close coordination with the SEC450 author and curriculum lead.
WHY 2026
2024.02
Arup HK$200M deepfake video BEC
2024.02
Microsoft/OpenAI joint state-actor LLM disclosure
2025.06
EchoLeak — zero-click M365 Copilot exfil
2025.08
ESET PromptLock — first LLM-runtime ransomware
DAY 5 — CAPSTONE
"For eight hours, you defend Verdancy Health against PROMETHEUS-7 — an AI-orchestrated adversary that has studied your AI SOC, knows how it reasons, and built an attack designed to make your own agents lie to you."
Verdancy Health Cooperative — a 14,000-employee regional healthcare insurer. The CISO publicly bragged about Verdancy's 'agentic AI SOC.' PROMETHEUS-7 — a financially motivated crew — scraped the talk and built the campaign around defeating that specific stack.
Hollow Mirror: Fintech
Halgrove Capital Partners · STYX-4
Hollow Mirror: OT
Brackenwell Industrial Systems · CINDERHOOK
Hollow Mirror: Public Sector
State of Lincoln DMV · PALEHORSE-9
Pass bar 700/1000 for GIAC capstone credit.
WHAT SHIPS
Day 1 content
Detector stack + AI phishing
Day 2 content
Deepfake BEC + workflow-gap detection
Day 3 content
LLM malware + EchoLeak + lethal trifecta
Day 4 content
Agentic adversaries + AI supply chain
Day 5 capstone
Operation Hollow Mirror — 8h immersive IR
Course blueprint
Full 5-day spec
MEET THE INSTRUCTOR
Domain authority
The intersection of AI tooling and network/security infrastructure is the exact technical territory this SEC5xx course covers — and the territory Ed has been building, teaching, and publishing in for the past three years.
Past delivery: graduate-level instruction at UPB (Bucharest) and conference-format workshops at industry events including AutoCon 5 (Munich, June 2026). Continuous practitioner work via vExpertAI consulting.
WHAT I'M ASKING FOR
01
Co-author with John Hubbard from blueprint forward.
02
Position as SEC450 follow-on in the curriculum map.
03
30-minute call to align on Day 1 module ownership.