position-lock.summary

Core Concept

Ethical reasoning ke liye aisi dilemmas par clear position lena zaroori hai jahan values conflict karti hain, us position ko evidence ke saath defend karna hota hai, aur honestly map karna hota hai ke kaun benefit karta hai aur cost kaun bear karta hai -- yeh woh kaam hai jo AI karne se bachta hai kyun ke woh balanced, non-committal answers default karta hai.

Key Mental Models

• Position Before Input: AI se consult karne se pehle apna stance writing mein commit karna aap ki reasoning ko AI ki diplomatic framing se shape hone se bachata hai -- sealed prediction lock independent thought prove karta hai
• Stakeholder Cost Matrix: Har ethical decision ke winners aur losers hote hain. Tamam affected groups (sirf sympathetic groups nahin) ko impact magnitude ke saath map karna aap ki position ki real cost reveal karta hai
• Calibrated Confidence: Confidence percentage ko specific reversal trigger ke saath pair karna genuine conviction ko stubbornness se alag karta hai

Critical Patterns

• Kisi bhi AI interaction se pehle timestamp ke saath apni position seal karein
• Sirf opinions nahin, evidence ke saath three distinct arguments likhein
• Har stakeholder group list karein, un groups ko bhi jinhein aap favor nahin karte
• Ek concrete, testable condition specify karein jo aapka mind change karegi

Common Mistakes

• Exercise ko "right answer" dhoondhne ke taur par treat karna jab ethical dilemmas deliberately single correct position nahin rakhtin
• Sirf woh stakeholder groups list karna jin ke saath aap sympathize karte hain aur uncomfortable impacts ignore karna
• Weak arguments ke saath confidence 95% rakhna, ya strong arguments ke saath 20% -- calibration ka matlab confidence ko evidence se match karna hai

Connections

• Builds on: Prediction Lock format (Chapter 1), systems thinking (Chapter 3), Confidence Calibration (Chapter 2)
• Leads to: Adversarial Defence (Exercise 2), Stakeholder Swap (Exercise 3), Decision Memo (Exercise 4)