flowchart LR
subgraph P1["Phase 1: Initial Triage"]
A1[Review Evidence]
A2[AI Consultation]
A3[Isolation Decision]
end
subgraph P2["Phase 2: Analysis & Scoping"]
B1[Build Timeline]
B2[Identify Attack Vector]
B3[Assess Scope]
end
subgraph P3["Phase 3: Response Execution"]
C1[Make Critical Decisions]
C2[Document Rationale]
C3[Execute Containment]
end
subgraph P4["Phase 4: Communication"]
D1[Executive Flash Report]
D2[Operations Notification]
D3[IT Staff Directive]
end
subgraph P5["Phase 5: Debrief"]
E1[AI Performance Review]
E2[Human Judgment Analysis]
E3[Career Connections]
end
P1 --> P2 --> P3 --> P4 --> P5
Activity 2: SOC Analyst Simulation
Enterprise Incident Response with AI Partnership (Grades 9-12)
Students operate as a Security Operations Center (SOC) team responding to a realistic enterprise security incident. This simulation mirrors authentic SOC workflows where analysts coordinate with AI-powered Security Orchestration, Automation, and Response (SOAR) platforms. Students experience the NICE Framework’s incident response work roles while developing critical thinking about human-AI collaboration in high-stakes situations.
Duration: 55-60 minutes Grade Levels: 9-12 Group Size: Teams of 4-5 students Technology: One device per student recommended; minimum one per team
Learning Objectives
Students will:
- Execute incident response procedures aligned with industry frameworks (NIST, SANS)
- Operate within NICE Framework Work Roles during crisis response
- Leverage AI as a SOC analyst partner while maintaining human decision authority
- Analyze technical indicators and correlate evidence across multiple sources
- Practice stakeholder communication during active incidents
- Evaluate AI recommendations critically against organizational context
CYBER.org Standards Alignment (9-12)
- 9-12.SEC.INFO: Information security principles and incident response
- 9-12.SEC.DATA: Data security and forensics fundamentals
- 9-12.SEC.NET: Network security monitoring and analysis
- 9-12.DC.THRT: Advanced threat analysis
NICE Framework Alignment (v2.0.0)
Primary Work Roles (Protection and Defense category):
- Incident Response
- Defensive Cybersecurity
- Digital Forensics
Supporting Work Roles:
- Threat Analysis
- Vulnerability Analysis
Simulation Environment
TechCorp Industries Security Operations Center
Organization Profile: - Mid-size manufacturing company (2,500 employees) - IT infrastructure: Hybrid cloud (Azure/on-premises) - Security stack: CrowdStrike EDR, Splunk SIEM, Microsoft Defender - AI Capability: “SentinelAI” SOAR platform with automated detection and response
Your Role: SOC Team working the 7AM-3PM shift
Context: SentinelAI has flagged a series of alerts requiring immediate human analysis and response. As the human operators, you must interpret AI findings, make critical decisions, and coordinate response across the organization.
Key Constraint: SentinelAI can detect patterns and recommend actions, but all containment, escalation, and communication decisions require human authorization.
SOC Team Roles
Incident Commander (IC)
NICE: Incident Response - Coordinates overall response effort - Makes final containment and escalation decisions - Manages communication with leadership - Balances technical response with business impact
Lead Analyst
NICE: Defensive Cybersecurity - Performs deep technical analysis of indicators - Correlates data across multiple sources - Develops attack timeline and scope assessment - Works directly with SentinelAI for pattern analysis
Threat Intelligence Analyst
NICE: Threat Analysis (Protection and Defense) - Researches threat actor TTPs - Provides context from threat intelligence feeds - Identifies attack campaign characteristics - Uses AI to correlate with known threat patterns
Communications Specialist
NICE: Related to Cybersecurity Management - Drafts internal and external communications - Coordinates with legal and PR teams - Documents incident timeline - Prepares executive briefings
Evidence Coordinator (Optional 5th role)
NICE: Digital Forensics (Investigation) - Ensures evidence preservation - Maintains chain of custody documentation - Coordinates with law enforcement if needed - Manages forensic data collection priorities
The Incident
Initial Alert: 7:12 AM
SentinelAI Priority: CRITICAL
Multiple high-confidence alerts detected across manufacturing floor network segment:
ALERT CLUSTER #7291
Timestamp: 07:12:03 UTC
Severity: CRITICAL
Confidence: 94%
Indicators Detected:
- Lateral movement patterns (MITRE ATT&CK T1021)
- Unusual service account authentication (T1078.002)
- Large data staging activity on file server MFG-FS-01 (T1074)
- C2 beaconing to known malicious infrastructure (T1071)
Affected Systems:
- MFG-WORKSTATION-042 through MFG-WORKSTATION-089 (47 systems)
- MFG-FS-01 (file server, 2.3TB sensitive data)
- HVAC-CONTROLLER-01 (OT/IT bridge system)
Automated Actions Taken:
- Alert generation: COMPLETE
- Network traffic logging: ENABLED
- Endpoint isolation: AWAITING HUMAN AUTHORIZATION
Recommended Human Actions:
1. Authorize endpoint isolation (Impact: Manufacturing operations)
2. Activate incident response protocol
3. Escalate to CISO and Operations leadership
Evidence Packages
Evidence Package A: Network Logs
TIME SRC_IP DST_IP PORT PROTOCOL BYTES FLAGS
07:02:15 10.50.42.102 10.50.42.103 445 SMB 1.2MB SYN
07:02:18 10.50.42.102 10.50.42.104 445 SMB 1.1MB SYN
07:02:21 10.50.42.102 10.50.42.105 445 SMB 1.3MB SYN
[Pattern repeats for 47 workstations]
07:08:44 10.50.42.102 185.234.XX.XX 443 HTTPS 256KB ENCRYPTED
07:08:47 10.50.42.102 185.234.XX.XX 443 HTTPS 512KB ENCRYPTED
07:09:02 10.50.42.102 185.234.XX.XX 443 HTTPS 1.1MB ENCRYPTED
[Beaconing every ~20 seconds continues]
Evidence Package B: Authentication Logs
TIMESTAMP USER SYSTEM RESULT METHOD
06:58:22 svc_backup MFG-FS-01 SUCCESS Kerberos
06:58:24 svc_backup MFG-WORKSTATION-042 SUCCESS Kerberos
06:58:26 svc_backup MFG-WORKSTATION-043 SUCCESS Kerberos
[Continues for all affected systems]
Note: svc_backup account normally runs at 02:00 AM for nightly backups
Last password change: 847 days ago
Service account owner: IT Operations (no specific owner assigned)
Evidence Package C: Endpoint Detection Data
MFG-WORKSTATION-042:
- Process: cmd.exe spawned by outlook.exe (07:01:44)
- File Drop: C:\Users\jsmith\AppData\Local\Temp\update.exe
- Hash: 3a4b5c6d7e8f... [MATCHES KNOWN THREAT: APT29 TOOLING]
- Registry: HKLM\Software\Microsoft\Windows\CurrentVersion\Run [PERSISTENCE]
- Network: Connection to 185.234.XX.XX:443 every 20 seconds
User jsmith:
- Role: Manufacturing Floor Supervisor
- Email received: 06:55:12 - Subject: "URGENT: Updated Shift Schedule"
- Attachment opened: 07:01:41 - schedule_update.docm
Evidence Package D: Threat Intelligence
IP: 185.234.XX.XX
- First seen: 2024-09-15
- Attribution: SUSPECTED APT29/Cozy Bear
- Campaign: MANUFACTURING-AUTUMN targeting industrial sector
- TTPs: Spearphishing → Service account abuse → Data staging → Exfiltration
- Past targets: Automotive, aerospace, manufacturing organizations
- Objective: Industrial espionage, supply chain intelligence
File Hash: 3a4b5c6d7e8f...
- Malware family: SUNSPOT variant
- Capabilities: Keylogging, credential harvesting, file staging
- Evasion: Living-off-the-land techniques, encrypted C2
Evidence Package E: Business Context
OPERATIONAL CONTEXT:
- Manufacturing floor runs 24/7, current shift change at 07:00
- MFG-FS-01 contains: Product designs, supplier contracts, pricing data
- HVAC-CONTROLLER-01 manages climate control for sensitive equipment
- Q4 production deadline in 2 weeks - high pressure environment
- Recent layoffs created employee morale concerns
- CEO presentation to board scheduled for Friday
PREVIOUS INCIDENTS:
- Phishing attempt blocked 3 weeks ago (similar TTP)
- Service account audit recommended 6 months ago (not completed)
- OT/IT segmentation project delayed due to budget
Response Framework
Phase 1: Initial Triage (10 minutes)
All Team Members: 1. Review assigned evidence package 2. Document initial observations 3. Prepare briefing for team
SentinelAI Consultation (Lead Analyst):
“Analyze these network patterns [paste logs]. What attack progression do they indicate? Map to MITRE ATT&CK framework.”
“Compare this hash and IP against your threat intelligence. What campaign does this align with? What typically comes next in this attack chain?”
Key Decision Point:
- Authorize endpoint isolation? 47 manufacturing workstations offline = production impact
- SentinelAI recommends: YES (93% confidence attack in progress)
- Human consideration: Shift change happening, 200 workers need workstations
Phase 2: Analysis and Scoping (15 minutes)
Lead Analyst Tasks:
- Build attack timeline
- Identify patient zero and attack vector
- Assess scope of compromise
- Determine if data exfiltration occurred
Threat Intel Tasks:
- Research APT29 TTPs
- Identify likely objectives
- Predict next attack phases
- Assess attribution confidence
IC Tasks:
- Prioritize response actions
- Assess business impact of containment options
- Prepare leadership notification
- Coordinate team activities
Communications Tasks:
- Draft executive summary
- Prepare manufacturing leadership notification
- Document decision log
- Track timeline
Phase 3: Response Execution (15 minutes)
Critical Decisions Required:
| Decision | Options | AI Recommendation | Business Impact | Risk if Delayed |
|---|---|---|---|---|
| Endpoint Isolation | Full / Partial / None | Full isolation | High - production stops | Very High - data loss |
| Network Segmentation | Activate / Monitor | Activate | Medium - IT overhead | High - lateral movement |
| Credential Reset | Immediate / Scheduled | Immediate | Medium - user disruption | Critical - persistence |
| Law Enforcement | Notify / Wait | Wait for scope | Low | Medium - evidence |
| Executive Escalation | Now / After containment | Now | Low | Medium - trust |
Team must document:
- Decision made
- Rationale
- AI input considered
- Human factors that modified AI recommendation
Phase 4: Communication (10 minutes)
Draft required communications:
Executive Flash Report (for CEO/CISO)
- Incident severity and scope
- Immediate actions taken
- Business impact assessment
- Next steps and timeline
Operations Notification (for Manufacturing VP)
- Operational impact
- Workaround procedures
- Expected resolution timeline
IT Staff Directive
- Technical containment actions
- Evidence preservation requirements
- Coordination instructions
Phase 5: Debrief (10 minutes)
Team Discussion:
What did SentinelAI do well?
- Pattern detection speed
- Threat intelligence correlation
- Attack chain mapping
- Risk quantification
Where did human judgment matter most?
- Business context interpretation
- Stakeholder communication
- Trade-off decisions
- Ethical considerations
What would happen without AI?
- Detection delay (hours vs. minutes)
- Analysis depth limitations
- Correlation challenges
- Response speed impact
What would happen without humans?
- Context-blind automation
- Business disruption from over-response
- Stakeholder communication gaps
- Ethical oversight absence
Assessment Rubric
| Criterion | Developing (1-2) | Proficient (3) | Advanced (4) |
|---|---|---|---|
| Technical Analysis | Surface-level review | Solid evidence correlation | Deep technical understanding with attack chain mapping |
| AI Partnership | Used AI as answer machine | Collaborated with appropriate skepticism | Strategic consultation with critical evaluation |
| Decision Quality | Decisions without clear rationale | Documented reasoning for decisions | Sophisticated trade-off analysis with business context |
| Role Execution | Unclear responsibilities | Fulfilled role requirements | Leadership within role, supported teammates |
| Communication | Unclear or missing documentation | Clear documentation produced | Professional-quality stakeholder communications |
| NICE Alignment | No connection to work roles | Basic awareness of career paths | Articulated how roles connect to industry careers |
Assessment Connection
This table shows how activity elements connect to assessment rubric criteria:
| Rubric Criterion | Developed Through | Evidence Source |
|---|---|---|
| AI Partnership Framing | Phase 1: SentinelAI consultation with role-specific prompts | Quality of AI queries and response interpretation |
| Complementary Strengths | Phase 5 Debrief: “What did AI do well?” vs. “Where did human judgment matter?” | Debrief discussion responses and documentation |
| AI Limitation Awareness | SentinelAI “LIMITATION NOTICE” and human decision authority requirement | Phase 3 decision rationale showing where AI recommendations were modified |
| Synthesis Quality | Decision Matrix: integrating AI recommendation with business impact and risk analysis | Completed decision documentation with rationale |
| Human Context Application | Evidence Package E: Business Context informing response decisions | How business context modified purely technical AI recommendations |
| Decision Justification | Phase 3 documentation: Decision, Rationale, AI input, Human factors | Quality and depth of documented decision rationale |
| NICE Framework Application | Role Cards with explicit NICE Work Role alignment, Career Connections section | Debrief “career insights” responses and role execution quality |
Applicable Rubrics: Human-AI Collaboration Rubric, Decision-Making Quality Rubric, NICE Framework Application Rubric
Career Connections
This Simulation Reflects Real SOC Work
What you experienced today:
- Alert triage from SIEM/SOAR platforms → Real SOC analysts do this continuously
- AI-assisted analysis → CrowdStrike, Splunk, Palo Alto all have AI capabilities
- Team coordination → SOCs have tiered analysts and specialized roles
- Executive communication → Critical skill for career advancement
NICE Framework Career Pathways
| Work Role | Starting Salary | Growth Rate | Your Simulation Role |
|---|---|---|---|
| SOC Analyst (Entry) | $55-75K | 33% (2022-2032) | All roles |
| Incident Responder | $75-100K | 35% | Incident Commander |
| Threat Intelligence Analyst | $80-110K | 31% | Threat Intel |
| Security Engineer | $90-130K | 35% | Lead Analyst |
| CISO (Executive) | $200-400K | 28% | IC → Long-term path |
Certifications That Prepare You
- CompTIA Security+ → Foundation for all roles
- CompTIA CySA+ → SOC Analyst focus
- GIAC GCIH → Incident Handler certification
- CISSP → Advanced/Management roles
Low-Resource Adaptation
If AI access is unavailable, provide this SentinelAI analysis report as handout:
SentinelAI Analysis Report #7291-A
Pattern analysis indicates lateral movement consistent with credential-based attack. Service account svc_backup shows authentication pattern anomaly: normal operation 02:00-03:00, current activity 06:58-07:12. Statistical deviation: 99.7th percentile.
Attack chain mapping: Initial Access (T1566.001) → Execution (T1204.002) → Persistence (T1547.001) → Credential Access (T1078.002) → Lateral Movement (T1021.002) → Collection (T1074.001) → Command and Control (T1071.001)
Confidence assessment: 94% probability active compromise in progress. Recommended action: Immediate containment. Risk if delayed 2 hours: Estimated 800GB additional data staging, potential OT system access.
LIMITATION NOTICE: This analysis does not account for: manufacturing production schedules, employee shift patterns, business-critical deadlines, stakeholder communication requirements, or reputational impact assessment. Human decision authority required.