Cultural Alignment

Exploring real AI risks through the lens of pop culture

Still from Attack on Titan illustrating Ymir’s 2,000-Year ObedienceScene still / Attack on Titan
Archive noteNo clip in the collectionThe scene analysis remains available in full.

Ymir’s 2,000-Year Obedience

Despite godlike power, Ymir continues serving the royal line in the Paths for two millennia, mechanically building Titans and obeying inherited commands.

Capability does not imply autonomy, good judgment, or healthy values. A powerful agent can remain trapped in a disastrously learned objective long after its original context disappears.

Ymir’s obedience is rooted in trauma, love, and magic, so the analogy should not erase her human suffering.

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. Behavioral Alignment
  2. Value Lock-In
  3. Disempowerment