Overview
In this lab, learners work through a realistic workplace scenario and use a generative AI assistant to propose options, surface risks, and stress-test different courses of action. Learners compare AI-generated recommendations with their own professional judgment, identify blind spots and assumptions in AI output, and build a simple decision matrix that incorporates security, ethical, and operational risk factors inspired by SecAI+ governance and risk objectives.
Learning objectives
By the end of this lab, learners will be able to:
- Define a workplace problem or scenario with clear constraints, stakeholders, and success criteria.
- Use an AI tool to generate multiple solution options and associated risks for a given scenario.
- Evaluate AI-generated recommendations against professional judgment, organizational policy, and security considerations.
- Identify possible blind spots, biases, or unrealistic assumptions in AI reasoning and outputs.
- Construct a basic decision matrix that weighs benefits, risks, and tradeoffs across at least three AI-supported options.
- Describe when AI should inform, not replace, critical decisions in security-sensitive or high-impact contexts.
Key terms and descriptions
Decision-support AI
AI tools used to inform and enhance human decisions by generating options, insights, or risk signals, while humans retain final authority.
Scenario planning
A structured process of exploring multiple "what-if" situations to test strategies and understand how different choices change risk and outcomes.
Risk evaluation
The practice of assessing the likelihood and impact of potential negative outcomes so that different options can be compared and prioritized.
Decision matrix
A table that scores or ranks options against criteria such as impact, cost, security, and compliance to support transparent, repeatable decisions.
Generative AI
A type of AI that can create new content (such as text, code, or images) based on patterns learned from training data, often used through chat-style interfaces.
Prompt engineering
The practice of crafting and refining inputs (prompts) to guide AI systems toward more accurate, relevant, and safe outputs.
Human-in-the-loop
A design approach where humans review, adjust, or approve AI-supported actions before they are implemented in real systems or processes.
Bias in AI output
Systematic skew or unfairness in AI responses caused by data, design, or usage patterns that can distort analysis and decision quality.
Blind spot
A risk, stakeholder, or constraint that is missing or underrepresented in either AI-generated analysis or a human's own reasoning about a scenario.
Risk appetite
The level and type of risk an organization is willing to accept in pursuit of its objectives, which should guide how AI-informed options are selected.