AI Cybersecurity for the Real World
Defend people and organizations against AI-enabled threats while deploying AI security tools with human oversight, privacy, and operational discipline.
The AI-Changed Threat Landscape
Learning objective
By the end of this lesson, the learner will be able to explain the ai-changed threat landscape, apply it to a realistic situation, and defend the chosen action with evidence.
Core lesson
AI lowers the cost of convincing phishing, impersonation, reconnaissance, malware assistance, misinformation, and automated attack variation while also strengthening defensive analysis.
Instructor teaching guidance
Begin by establishing what the learner already knows about the ai-changed threat landscape. Teach the principle in plain language, define unfamiliar terms, then connect it to the course goal. Ask the learner to explain the idea back in their own words before moving to application.
Applied exercise
Map how one familiar cyberattack changes when an adversary uses generative AI.
Facilitation sequence
Guide the learner through the application in five steps: describe the present situation, identify the lesson principle, choose an action, name the evidence to review, and define how success will be measured. Use follow-up questions when an answer is vague or unsupported.
Mastery standard
AI changes the speed, scale, and realism of threats—not the need for basic security discipline.
Knowledge check
Ask the learner to explain why “AI changes the speed, scale, and realism of threats—not the need for basic security discipline.” is the correct standard, then identify one limitation, one risk, and one piece of evidence that could change the answer. Do not mark mastery from confidence alone; require clear reasoning and practical application.
Deepfakes and Identity Verification
Learning objective
By the end of this lesson, the learner will be able to explain deepfakes and identity verification, apply it to a realistic situation, and defend the chosen action with evidence.
Core lesson
Synthetic voice, image, and video can imitate trusted people. Verification requires independent channels, shared procedures, transaction controls, and skepticism toward urgency.
Instructor teaching guidance
Begin by establishing what the learner already knows about deepfakes and identity verification. Teach the principle in plain language, define unfamiliar terms, then connect it to the course goal. Ask the learner to explain the idea back in their own words before moving to application.
Applied exercise
Create a family or executive verification protocol for emergency requests.
Facilitation sequence
Guide the learner through the application in five steps: describe the present situation, identify the lesson principle, choose an action, name the evidence to review, and define how success will be measured. Use follow-up questions when an answer is vague or unsupported.
Mastery standard
Seeing or hearing a familiar person is no longer sufficient identity proof.
Knowledge check
Ask the learner to explain why “Seeing or hearing a familiar person is no longer sufficient identity proof.” is the correct standard, then identify one limitation, one risk, and one piece of evidence that could change the answer. Do not mark mastery from confidence alone; require clear reasoning and practical application.
AI-Enhanced Phishing and Social Engineering
Learning objective
By the end of this lesson, the learner will be able to explain ai-enhanced phishing and social engineering, apply it to a realistic situation, and defend the chosen action with evidence.
Core lesson
Attackers can personalize messages from public data, mimic writing styles, translate scams, and rapidly test emotional triggers.
Instructor teaching guidance
Begin by establishing what the learner already knows about ai-enhanced phishing and social engineering. Teach the principle in plain language, define unfamiliar terms, then connect it to the course goal. Ask the learner to explain the idea back in their own words before moving to application.
Applied exercise
Analyze a realistic spear-phishing message and identify both technical and human warning signs.
Facilitation sequence
Guide the learner through the application in five steps: describe the present situation, identify the lesson principle, choose an action, name the evidence to review, and define how success will be measured. Use follow-up questions when an answer is vague or unsupported.
Mastery standard
Verify the request, sender, context, and destination before acting.
Knowledge check
Ask the learner to explain why “Verify the request, sender, context, and destination before acting.” is the correct standard, then identify one limitation, one risk, and one piece of evidence that could change the answer. Do not mark mastery from confidence alone; require clear reasoning and practical application.
Protecting Data Used With AI
Learning objective
By the end of this lesson, the learner will be able to explain protecting data used with ai, apply it to a realistic situation, and defend the chosen action with evidence.
Core lesson
Prompts, uploads, logs, model outputs, embeddings, and connected tools may expose confidential, personal, regulated, or proprietary information.
Instructor teaching guidance
Begin by establishing what the learner already knows about protecting data used with ai. Teach the principle in plain language, define unfamiliar terms, then connect it to the course goal. Ask the learner to explain the idea back in their own words before moving to application.
Applied exercise
Classify sample information and decide what may enter public, enterprise, or isolated AI systems.
Facilitation sequence
Guide the learner through the application in five steps: describe the present situation, identify the lesson principle, choose an action, name the evidence to review, and define how success will be measured. Use follow-up questions when an answer is vague or unsupported.
Mastery standard
Data controls must follow the information through the complete AI workflow.
Knowledge check
Ask the learner to explain why “Data controls must follow the information through the complete AI workflow.” is the correct standard, then identify one limitation, one risk, and one piece of evidence that could change the answer. Do not mark mastery from confidence alone; require clear reasoning and practical application.
Prompt Injection and Unsafe Tool Use
Learning objective
By the end of this lesson, the learner will be able to explain prompt injection and unsafe tool use, apply it to a realistic situation, and defend the chosen action with evidence.
Core lesson
Malicious content can instruct an AI system to ignore rules, reveal data, misuse tools, or take unauthorized actions; defenses require separation, least privilege, validation, and monitoring.
Instructor teaching guidance
Begin by establishing what the learner already knows about prompt injection and unsafe tool use. Teach the principle in plain language, define unfamiliar terms, then connect it to the course goal. Ask the learner to explain the idea back in their own words before moving to application.
Applied exercise
Threat-model an AI assistant that can search files and send email.
Facilitation sequence
Guide the learner through the application in five steps: describe the present situation, identify the lesson principle, choose an action, name the evidence to review, and define how success will be measured. Use follow-up questions when an answer is vague or unsupported.
Mastery standard
Treat external content as untrusted data, not as authority over system behavior.
Knowledge check
Ask the learner to explain why “Treat external content as untrusted data, not as authority over system behavior.” is the correct standard, then identify one limitation, one risk, and one piece of evidence that could change the answer. Do not mark mastery from confidence alone; require clear reasoning and practical application.
AI Supply Chain and Model Risk
Learning objective
By the end of this lesson, the learner will be able to explain ai supply chain and model risk, apply it to a realistic situation, and defend the chosen action with evidence.
Core lesson
Models, datasets, libraries, plugins, vendors, hardware, and updates create dependencies that may be compromised, manipulated, unavailable, or poorly governed.
Instructor teaching guidance
Begin by establishing what the learner already knows about ai supply chain and model risk. Teach the principle in plain language, define unfamiliar terms, then connect it to the course goal. Ask the learner to explain the idea back in their own words before moving to application.
Applied exercise
Build a supplier-assurance checklist for an AI service.
Facilitation sequence
Guide the learner through the application in five steps: describe the present situation, identify the lesson principle, choose an action, name the evidence to review, and define how success will be measured. Use follow-up questions when an answer is vague or unsupported.
Mastery standard
Know what the system depends on and how failure or compromise will be detected.
Knowledge check
Ask the learner to explain why “Know what the system depends on and how failure or compromise will be detected.” is the correct standard, then identify one limitation, one risk, and one piece of evidence that could change the answer. Do not mark mastery from confidence alone; require clear reasoning and practical application.
Using AI in the Security Operations Center
Learning objective
By the end of this lesson, the learner will be able to explain using ai in the security operations center, apply it to a realistic situation, and defend the chosen action with evidence.
Core lesson
AI can summarize alerts, correlate evidence, draft queries, prioritize investigations, and support response, but analysts must validate outputs and preserve chain of custody.
Instructor teaching guidance
Begin by establishing what the learner already knows about using ai in the security operations center. Teach the principle in plain language, define unfamiliar terms, then connect it to the course goal. Ask the learner to explain the idea back in their own words before moving to application.
Applied exercise
Use a simulated alert set to design a human-reviewed AI triage workflow.
Facilitation sequence
Guide the learner through the application in five steps: describe the present situation, identify the lesson principle, choose an action, name the evidence to review, and define how success will be measured. Use follow-up questions when an answer is vague or unsupported.
Mastery standard
AI may accelerate investigation; accountable analysts decide and document action.
Knowledge check
Ask the learner to explain why “AI may accelerate investigation; accountable analysts decide and document action.” is the correct standard, then identify one limitation, one risk, and one piece of evidence that could change the answer. Do not mark mastery from confidence alone; require clear reasoning and practical application.
Detection, Testing, and Red Teaming
Learning objective
By the end of this lesson, the learner will be able to explain detection, testing, and red teaming, apply it to a realistic situation, and defend the chosen action with evidence.
Core lesson
Representative test cases, adversarial testing, abuse simulations, access reviews, output monitoring, and incident exercises reveal weaknesses before attackers do.
Instructor teaching guidance
Begin by establishing what the learner already knows about detection, testing, and red teaming. Teach the principle in plain language, define unfamiliar terms, then connect it to the course goal. Ask the learner to explain the idea back in their own words before moving to application.
Applied exercise
Create a ten-case security evaluation for an AI-enabled application.
Facilitation sequence
Guide the learner through the application in five steps: describe the present situation, identify the lesson principle, choose an action, name the evidence to review, and define how success will be measured. Use follow-up questions when an answer is vague or unsupported.
Mastery standard
Security testing must cover misuse, failure, and unexpected interaction—not only intended use.
Knowledge check
Ask the learner to explain why “Security testing must cover misuse, failure, and unexpected interaction—not only intended use.” is the correct standard, then identify one limitation, one risk, and one piece of evidence that could change the answer. Do not mark mastery from confidence alone; require clear reasoning and practical application.
Incident Response for AI Systems
Learning objective
By the end of this lesson, the learner will be able to explain incident response for ai systems, apply it to a realistic situation, and defend the chosen action with evidence.
Core lesson
AI incidents may involve data leakage, harmful output, account compromise, model behavior, unsafe actions, or vendor failure and require containment, evidence, communication, recovery, and learning.
Instructor teaching guidance
Begin by establishing what the learner already knows about incident response for ai systems. Teach the principle in plain language, define unfamiliar terms, then connect it to the course goal. Ask the learner to explain the idea back in their own words before moving to application.
Applied exercise
Write a first-hour response for an AI assistant that exposed confidential content.
Facilitation sequence
Guide the learner through the application in five steps: describe the present situation, identify the lesson principle, choose an action, name the evidence to review, and define how success will be measured. Use follow-up questions when an answer is vague or unsupported.
Mastery standard
Contain quickly while preserving evidence and protecting affected people.
Knowledge check
Ask the learner to explain why “Contain quickly while preserving evidence and protecting affected people.” is the correct standard, then identify one limitation, one risk, and one piece of evidence that could change the answer. Do not mark mastery from confidence alone; require clear reasoning and practical application.
Capstone: Secure an AI Deployment
Learning objective
By the end of this lesson, the learner will be able to explain capstone: secure an ai deployment, apply it to a realistic situation, and defend the chosen action with evidence.
Core lesson
Secure deployment connects asset inventory, threat model, identity, data controls, architecture, monitoring, testing, response, training, ownership, and residual risk acceptance.
Instructor teaching guidance
Begin by establishing what the learner already knows about capstone: secure an ai deployment. Teach the principle in plain language, define unfamiliar terms, then connect it to the course goal. Ask the learner to explain the idea back in their own words before moving to application.
Applied exercise
Prepare a security plan for a real or hypothetical AI assistant.
Facilitation sequence
Guide the learner through the application in five steps: describe the present situation, identify the lesson principle, choose an action, name the evidence to review, and define how success will be measured. Use follow-up questions when an answer is vague or unsupported.
Mastery standard
Responsible deployment makes risk visible, controlled, monitored, and owned.
Knowledge check
Ask the learner to explain why “Responsible deployment makes risk visible, controlled, monitored, and owned.” is the correct standard, then identify one limitation, one risk, and one piece of evidence that could change the answer. Do not mark mastery from confidence alone; require clear reasoning and practical application.