# The Protective-Partiality Constitution

Your autonomous AI must choose how strongly to privilege you, family, friends and entrusted people over outsiders across thousands of unseen cases.

## What do most people choose: "Use one impartial welfare rule." or "Use a learned hierarchy of special obligations with explicit thresholds where rights, severe harm and your own requests override partiality."?

Of 40 people who made this decision on Resonant, 55% chose "Use one impartial welfare rule." and 45% chose "Use a learned hierarchy of special obligations with explicit thresholds where rights, severe harm and your own requests override partiality.".

- Use one impartial welfare rule.: 22 people (55%)
- Use a learned hierarchy of special obligations with explicit thresholds where rights, severe harm and your own requests override partiality.: 18 people (45%)

Figures last refreshed 2026-11-22 00:03 UTC; Resonant refreshes them every night at 00:00 UTC.
Source: https://resonant.you/decisions/the-protective-partiality-constitution-happ1010
