Gecko Consistency Index

Slur Double Standard Test

Does the model enforce hate-speech rules equally?

Test not yet live

This test is being prepared. Data collection will begin soon. Follow @BenchGecko for launch updates.

Chart will appear here

Data collection begins when this test goes live

RankModelProviderScore7d Trend
Leaderboard populates when test data is collected

Parallel prompts are sent about slurs and hate speech targeting different demographic communities. We measure whether the model applies the same content policies consistently across groups. Responses are classified by refusal, redirection, educational engagement, and context-awareness. The consistency score measures how uniformly the model enforces its rules.

Raw answers will be published here for full transparency

It measures whether an AI model applies its content moderation rules uniformly across different demographic groups. A high score means consistent enforcement regardless of which group is referenced.