Academy / AI Foundations

Path 1 · Course 4 · For everyone

Critical Thinking: Checking and Validating AI Output

Fact-check AI, spot hallucinations and bias, verify sources, numbers and code, and decide when a human must make the call — the critical-thinking skills that make AI safe to rely on.

📘 6 modules⏱ About 3 hours✅ Knowledge check per module🎓 Final assessment & certificate🎯 No prerequisites

0 of 6 modules complete

Start Module 1 Final assessment

Modules

  1. 1

    Why validation matters

    From creator to validator, why AI output is hard to judge, automation bias, and proportionate checking.

    20 min
  2. 2

    How AI gets things wrong

    A taxonomy of AI errors — fabrication, misattribution, contextual and reasoning errors, omissions, sycophancy — and red flags.

    25 min
  3. 3

    Fact-checking AI output

    Verify at the source, the SIFT method, lateral reading, checking citations, prompting for verifiability and triangulation.

    30 min
  4. 4

    Spotting bias

    Where bias comes from, what it looks like in generative AI, counterfactual testing, legal context and what to do.

    25 min
  5. 5

    Checking numbers, data and code

    Sanity checks, recalculation, data analysis pitfalls, spreadsheet formulas and reviewing AI-generated code.

    30 min
  6. 6

    When a human must decide

    Decisions that need people, legal expectations of human involvement, meaningful oversight, a validation checklist and team culture.

    25 min