🤔 Neuroethical Decision Framework
עקרון: Show WHY each AI recommendation. Opt-out is default. Run ML vs family-preferred in parallel. Let humans decide.
Current AI Recommendation:
Action: Increase transfer matrix boosters
Confidence: 87%
Reason: Last 4 weeks show 60% skill retention decay. Weekly boosters predicted to restore 80% by week 6.
Transparency Pipeline
Family accepts AI recommendation
✅ Explained rationale + showed data
Family rejects (wants different path)
✅ Run both in parallel. Compare outcomes.
Family wants no AI involvement
✅ Opt-out works. Manual family-chosen plan only.
🎯 Dual-Track Comparison:
If family rejects AI recommendation: run ML-recommended path + family-preferred path in parallel. Track which wins. Learn trust over time.
