---
aliases:
  en:
  - CEV
concept_id: coherent_extrapolated_volition
preferred:
  en: Coherent Extrapolated Volition
  ja: 首尾一貫した外挿的意志
relations:
  relations:
    broader:
    - alignment_target
    prerequisite:
    - human_preferences
    - human_values
    related:
    - friendly_ai
status: active
---

# Coherent Extrapolated Volition

A historical proposal for an alignment target in which an advanced AI would act on the parts of humanity's preferences that would converge under an idealized extrapolation in which people knew more, thought faster, better understood themselves, and had more opportunity to deliberate together.

## broader

- [alignment target](../terminology/alignment_target.txt)

## prerequisite

- [human preferences](../terminology/human_preferences.txt)
- [human values](../terminology/human_values.txt)

## related

- [Friendly AI](../terminology/friendly_ai.txt)

## Documents: about

- [CEV 型 Sovereign と訂正可能な AGI は異なる安全化戦略である](../documents/yudkowsky.cev-sovereign-vs-corrigible-agi.txt)

[Corpus index](../index.txt)