AI safety concept
AI Interpretability and Transparency
Interpretability aims to explain how an AI reaches decisions so people can spot errors, hidden goals, and unsafe behavior.
Scenario file
Scenes in this index
13Sort scenes

3 Body Problem2024
The Wallfacer Project

Contact1997
Eighteen Hours of Static

Lost2006
Push the Button

The Twilight Zone1963
The Old Man in the Cave

The Prestige2006
The Duplicate Tanks

The Hitchhiker’s Guide to the Galaxy1981
The Answer Is 42

The Zone of Interest2023
Vines Cover the Wall

Twin Peaks1990
Cooper Cannot Retrieve the Answer

Toy Story1995
Buzz’s Lucky Flight

Anatomy of a Fall2023
The Tape Is Not the Marriage

True Detective2014
The Green-Eared Monster

Men in Black1997
The Little Tiffany Test

The Wizard of Oz1939
Toto Pulls Back the Curtain