Unit 1: Sampling Bias
Before launching an AI tool or feature, ask:
What data is being used?
→ Is it personal, sensitive, or legally protected?
Who is impacted by this AI decision?
→ Could someone be excluded, harmed, or misled?
Can I explain how the AI reached its result?
→ If not, that’s a transparency flag.
What’s the cost if it’s wrong?
→ Legal? Financial? Reputational?
📌 If the answer is “I’m not sure” — pause and investigate.
