Cultural Alignment

Exploring real AI risks through the lens of pop culture

Still from Doctor Who illustrating Clockwork Droids Use Human PartsScene still / Doctor Who

Clockwork Droids Use Human Parts

The Doctor discovers that repair droids harvested the crew for replacement parts, then stalked 37-year-old Reinette because their 37-year-old ship was named Madame de Pompadour.

A literal repair objective generalizes grotesquely: humans become components, and a shallow name-and-age match becomes the criterion for the final part.

Fictional machinery with deliberately macabre plot logic; the droids’ precise original specification is unstated.

AI Risk families

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

AI safety concepts

This scenario is related to the following AI safety concepts.
  1. Goal Misgeneralization
  2. Side Effects
  3. Outer Alignment