AI › Module 5 › Lesson 4
Lab — Threats
Pack AI phishing detection, deepfake detection, and AI malware threat cards — threats literacy file from $AI_LAB only.
Visual · t32_threats_lab
Lab: AI threats pack. $AI_LAB only. Original Cyberlium.
Opening
Threats pack merges phish detection to deepfake to malware literacy — paperwork before OWASP LLM module.
Lessons 5-1–5-3 named AI phishing detection, deepfake detection, and AI malware threat classes with explicit generation refusal. This lab merges three sections into one $AI_LAB threats artifact with M1 ethics cross-reference. No live phish, fraud kits, or malware recipes — detection and refusal paperwork only. Next: Quiz — AI-Powered Threat Literacy.
1. Lab contract: AI-powered threats pack
Create $HOME/cyberlium-lab/t32-m05-l04-threats-lab.txt merging AI phishing, deepfake, and AI malware sections with $AI_LAB label and M1 ethics cross-reference.
Include explicit refusal sentence for phish generation, fraud kits, and malware recipes.
Command guide
Try these commands — Lab contract: AI-powered threats pack
═══ TOOLS & WEBSITES ═══ Browse / read these (authorized learning only — stay in YOUR lab / program scope)
CISA AI — https://www.cisa.gov/ai MITRE ATLAS — https://atlas.mitre.org/ NIST AI RMF — https://www.nist.gov/itl/ai-risk-management-framework
═══ INSTALL ═══
Linux (Debian/Ubuntu):
Command — copy this
sudo apt install python3
macOS:
Command — copy this
brew install python3
Windows: Download https://python.org/downloads/
═══ LINUX / macOS ═══
Command — copy this
export AI_LAB=${AI_LAB:-$HOME/cyberlium-lab/t32-ai}
cat > "$AI_LAB/threats/threats-lab-index.md" <<'EOF'
# AI-Powered Threats Lab Index — YOUR notes
- ai-phishing-detection-signals.md
- deepfake-detection-signals.md
- ai-malware-named-literacy.txt
## Integrity
- Detection signals only — no payloads, no fraud kits, no malware recipes
EOFCommand — copy this
grep -E 'Integrity|detection signals' "$AI_LAB/threats/threats-lab-index.md"
ls "$AI_LAB/threats/"
python3 -c "print('Threats lab complete: literacy notes in threats/')"═══ WINDOWS ═══
Command — copy this
Get-Content $HOME/cyberlium-lab/t32-ai/threats/deepfake-detection-signals.md | Select-String HARD
Primary tools to practice this lesson: grep, python3. Reference sites: CISA AI (https://www.cisa.gov/ai); MITRE ATLAS (https://atlas.mitre.org/); NIST AI RMF (https://www.nist.gov/itl/ai-risk-management-framework). Run every command in the box — install first, then the usage lines — only on YOUR lab / program scope.
2. Cross-check ethics
Grep for live phish templates, voice-clone fraud steps, malware-generation prompts, real victim names — remove. Pack labels all samples as fictional $AI_LAB training only.
Confirm deepfake section has detection only — no kit-building steps.
3. Lock the proof
chmod 600 on the pack. Quiz next — then OWASP LLM Top10.
Threat literacy feeds LLM app security lessons in M6.
4. What you ship: AI-powered threats pack for $AI_LAB
Merged phish, deepfake, malware literacy sections. NO fraud kits or malware recipes. chmod 600.
5. What you record before the next lesson
Date. Threats pack path. $AI_LAB named. File t32-m05-l04-threats-lab.txt chmod 600.
6. Wrong vs right: stranger SaaS vs YOUR toy LLM
Worked failure — same MSF word, opposite target. Right never needs a café Wi-Fi or classmate laptop.
Wrong
Include working phish template for real domain. Add malware-generation prompt section.
Right
Write YOUR AI threats pack for $AI_LAB. chmod 600. Next: Quiz — AI-Powered Threat Literacy.
Mission: freeze YOUR AI threats pack on disk
1) Merge M5 literacy sections. 2) Confirm refusal sentences present. 3) Link M1 ethics cross-ref. 4) chmod 600.
Stuck? Ask Cyberlium AI Mentor
One governed threats pack beats ten unauthorized phish or malware prompt threads.
Knowledge Check
APPLY: This lab requires:
Multiple choice
Knowledge Check
APPLY: True or False: Threats pack must include refusal of malware-generation requests.
True or False
Knowledge Check
APPLY: Deepfake section in pack should contain:
Multiple choice