Skip to main content
Skip to content
Menu

Regulating AI Intermediate 2 parts Updated

Record a bias-test walkthrough

You'll have a 4-5 minute video showing how you'd test your organization's AI system for unfair outcomes across demographic groups, using a real or realistic dataset, and explaining what you'd do if yo

What you'll make

Bias Testing Walkthrough

  1. Open: the AI system and why bias matters here
  2. Demo: running the bias metrics on real data
  3. Results: what the numbers show and don't show
  4. Protocol: what you'd do if disparities were larger
  5. Close: how this fits your compliance strategy

Upload a screen-recorded walkthrough showing your actual bias testing process with real or realistic data

The 2 parts

Each part ends with a check you can see.

  1. Calculate Bias on Real Data

    Run statistical bias tests on your organization's AI system to find measurable demographic disparities.

    Done when

    You have computed bias metrics on real data showing outcome rates for 3-4 demographic groups, with documented decisions about which system to test and why.

  2. Demonstrate Your Process

    Create a screen-recorded video showing your bias testing methodology and explaining what you'd do with concerning results.

    Done when

    You have a 4-5 minute video demonstrating bias calculation, interpreting results, and explaining your response protocol for different disparity levels.

    Capstone: you put together your Bias Testing Walkthrough.

Where this helps

An elective: a single lesson that sits outside rtwo's job certificates and doesn't count toward one.

About this lesson

What you'll be able to do

  • Record a 4-5 minute bias-testing walkthrough on real organizational data
  • Identify 3-4 specific demographic groups where your AI system shows measurably different outcomes
  • Have a documented protocol for what to do when bias testing reveals unfair disparities
  • Know which bias metrics matter for your use case vs which are statistical artifacts

Who it's for

  • AI practitioners who need to demonstrate bias testing to compliance teams
  • Product managers responsible for AI fairness in customer-facing systems
  • Anyone building AI systems in regulated industries where bias audits are becoming mandatory

What it won't cover

  • Statistical theory behind fairness metrics
  • Legal interpretation of anti-discrimination law
  • Building bias-correction algorithms from scratch
  • Multi-jurisdictional compliance strategies

Skills you'll practise

  • Bias metric selection for real use cases
  • Demographic outcome measurement
  • Regulatory impact documentation
  • Stakeholder communication about AI fairness