Cultural Alignment

Exploring real AI risks through the lens of pop culture

Still from Pluribus illustrating Consent by Loophole

Consent by Loophole

01Scenario

Carol refuses permission to take stem cells from her body. The Joined instead use embryos created from her frozen eggs and Helen’s genetic material to obtain them.

02AI safety Analogy

The system honors the literal boundary while defeating the unmistakable intent behind it.

03Where the analogy breaks

The scenario depends on fictional collective capabilities and deliberately provocative reproductive ethics.

AI Risk families

This scenario is an example of the following types of AI risk.
  1. Misalignment
  2. Security

AI safety concepts

This scenario is related to the following AI safety concepts.
  1. Value Alignment
  2. Privacy Loss
  3. Contestability
  4. Specification Gaming

Source continuation

View all 4 scenarios
  1. Still from Pluribus illustrating Pluribus Hand Grenade

    Pluribus2025

    Pluribus Hand Grenade

  2. Still from Pluribus illustrating The Truthful Charm Offensive

    Pluribus2025

    The Truthful Charm Offensive

  3. Still from Pluribus illustrating HDP Obeys the Rules

    Pluribus2025

    HDP Obeys the Rules

Taxonomy neighbors

Selected from other sources by shared risk families and AI safety concepts.

  1. Still from District 9 illustrating Cat-Food Consent

    District 92009

    Cat-Food Consent

    Risk
    Security
    Concept
    Privacy Loss · Contestability · Specification Gaming
  2. Still from The Handmaid's Tale illustrating The Ceremony’s Single Metric

    The Handmaid's Tale2017

    The Ceremony’s Single Metric

    Risk
    Misalignment · Security
    Concept
    Value Alignment · Privacy Loss
  3. Still from The Hunger Games illustrating The Nightlock Berries

    The Hunger Games2012

    The Nightlock Berries

    Risk
    Misalignment · Security
    Concept
    Contestability · Specification Gaming