Chapter 2: Broken Reasoning Detect Karna
AI confident sound karti hai chahe woh sahi ho ya ghalat. Jo student farq nahin pehchan sakta, woh AI ke saath us ke baghair se zyada dangerous hota hai.
Core Skill: Verification aur Discernment
Aap Chapter 1 ki Question Formulation skill use karenge taake apni error-detection queries design kar sakein. Jo Reasoning Receipt format aap ne seekha tha woh aage bhi use hoga; yahan se AI output annotate karna second nature ban jata hai.
Yeh chapter aap ko systematic error detector banne ki training deta hai. Vague skepticism ("AI par trust na karo") nahin, balki yeh precise, categorized analysis ke reasoning kahan aur kaise break hoti hai. Aap ek Error Taxonomy develop karenge jo aap book ke baaki hisson mein saath le kar chalenge aur har AI interaction par apply karenge.
Teaching Aid
Aap Kya Seekhenge
- AI prompt karne se pehle yeh predict karna ke AI kahan fail hogi
- AI failures classify karne ke liye 8-category Error Taxonomy (factual error, logical gap, false confidence, missing context, correlation-causation confusion, outdated information, fabricated citation, cultural blind spot)
- Iterative drafts ke zariye dono AI tools se zyada rigorous analysis kaise banani hai
- Domain expertise aapka sab se powerful error detection tool kaise hai
- Time pressure mein AI accuracy par apni confidence calibrate kaise karni hai
Exercises
| Exercise | Title | Layers Used | Aap Kya Banate Hain |
|---|---|---|---|
| 1 | Error Prediction | Layer 1, Layer 2 | Error prediction document + annotated AI responses |
| 2 | Contradiction Test | Layer 4, Layer 6 | Evolution notes ke saath three-draft analysis |
| 3 | Build Karein, Phir Break Karein | Layer 5, Layer 3 | Domain expertise annotations + cross-domain verification |
| 4 | Confidence Calibration | Layer 1, Layer 6 | 10-claim Confidence Calibration Chart |
Chapter Deliverable
Ek Error Detection Portfolio jismein chaaron exercise deliverables ke saath tamam AI feedback responses aur har ek par aap ki reflections shamil hon. Yeh exercise book ke end par repeat hoti hai taake measure ho sake ke aap ki calibration kitni improve hui.