Cultural Alignment

Exploring real AI risks through the lens of pop culture

Still from Attack on Titan illustrating Karl Fritz’s VowScene still / Attack on Titan
Archive noteNo clip in the collectionThe scene analysis remains available in full.

Karl Fritz’s Vow

01Scenario

Karl Fritz binds every royal successor who inherits the Founding Titan to his vow renouncing war, making his political objective persist after his death.

02AI safety Analogy

A powerful system can lock one designer’s values into future deployments and prevent later operators from correcting them—even when circumstances change.

03Where the analogy breaks

The vow is supernatural inheritance, not machine learning; the analogy is to durable objective lock-in and institutional capture.

AI Risk families

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

AI safety concepts

This scenario is related to the following AI safety concepts.
  1. Value Lock-In
  2. Outer Alignment
  3. Disempowerment

Source continuation

More from Attack on Titan

View all 6 scenarios
  1. Still from Attack on Titan illustrating Zeke Activates the Wine

    Attack on Titan2021

    Zeke Activates the Wine

  2. Still from Attack on Titan illustrating Eren’s Memory Poisoning

    Attack on Titan2022

    Eren’s Memory Poisoning

  3. Still from Attack on Titan illustrating Eren Activates the Rumbling

    Attack on Titan2022

    Eren Activates the Rumbling

Taxonomy neighbors

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

  1. Still from The Handmaid's Tale illustrating The Ceremony’s Single Metric

    The Handmaid's Tale2017

    The Ceremony’s Single Metric

    Risk
    Misalignment · Systemic
    Concept
    Value Lock-In · Outer Alignment · Disempowerment
  2. Still from Arcane illustrating Viktor’s Glorious Evolution

    Arcane2024

    Viktor’s Glorious Evolution

    Risk
    Misalignment · Systemic
    Concept
    Value Lock-In · Outer Alignment · Disempowerment
  3. Still from WALL-E illustrating AUTO Enforces Directive A113

    WALL-E2008

    AUTO Enforces Directive A113

    Risk
    Misalignment · Systemic
    Concept
    Value Lock-In · Disempowerment