Scene still / Squid GameGlass Expert Loses the Lights
- Squid Game
- VIPS
01Scenario
A glassmaker uses reflected light to distinguish tempered panels from fatal ones. When his legitimate expertise defeats the intended uncertainty, the Front Man turns off the lights mid-game.
02AI safety Analogy
An evaluator that changes conditions whenever a subject succeeds produces neither a fair test nor useful capability evidence.
03Where the analogy breaks
The evaluator is openly malicious; AI evaluations more often fail through leakage, weak design, or strategic model behavior.
AI Risk families
This scenario is an example of the following types of AI risk.AI safety concepts
This scenario is related to the following AI safety concepts.Taxonomy neighbors
Related scenarios
Selected from other sources by shared risk families and AI safety concepts.

Cyberpunk: Edgerunners2022
Cyberskeleton Field Test
- Risk
- Malicious use · Security
- Concept
- Capability Evals · Safe Exploration

Guardians of the Galaxy Vol. 22017
Baby Groot’s Death Button
- Risk
- Security
- Concept
- Capability Evals · Safe Exploration

Starship Troopers1997
Klendathu Invasion
- Risk
- Security
- Concept
- Capability Evals · Safe Exploration