
Create a decision-log of moments when you almost over-relied on AI recommendations
You'll have a 5-10 entry log identifying patterns of when you were tempted to accept AI guidance without pushback, with notes on what independent verification revealed.
What you'll make
AI Over-Reliance Decision Log
- Decision Moment: context and AI recommendation
- Near-Acceptance: why it felt convincing
- Independent Check: what verification revealed
- Pattern Recognition: recurring bias types
Upload a screenshot of your completed log showing decision moments, AI recommendations, and what independent verification revealed
The 4 parts
Each part ends with a check you can see.
Test AI's Consensus Pull
Compare how different AI models respond to the same controversial question from your field.
Done when
You have tested one work-relevant controversial question across 3+ AI models and documented where they converge vs diverge
Document Near-Acceptance Moments
Log real moments over one week when you almost accepted AI recommendations without verification.
Done when
You have 3-5 documented moments when AI recommendations felt convincing enough that you almost acted on them without checking
Run Independent Verification
Take your near-acceptance moments and verify each recommendation against non-AI sources.
Done when
You have verified 3+ AI recommendations through independent sources and documented what verification revealed
Identify Your Bias Patterns
Analyze your decision log to find systematic patterns in when you're most vulnerable to over-reliance.
Done when
You can name 2+ specific conditions when you're most likely to accept AI recommendations without verification, with evidence from your log
Capstone: you put together your AI Over-Reliance Decision Log.
About this lesson
What you'll be able to do
- Identify 5+ moments when you almost accepted AI recommendations without verification
- Test AI responses against your own judgment to spot systematic bias
- Maintain your own perspective when AI outputs feel authoritative
- Create a reusable process for catching over-reliance before it happens
Who it's for
- Professionals using AI for decisions or analysis who want to avoid groupthink
- People who notice AI answers feel convincing but wonder what they're missing
- Anyone building AI into their workflow who needs to stay independently critical
What it won't cover
- How to prompt AI better
- Technical details of model training
- Political analysis of AI company policies
- Building custom AI models
Skills you'll practise
- Critical evaluation of AI recommendations
- Bias detection through comparative testing
- Independent verification methods
- Decision-making under AI influence