Question 1

Which AI model censors the most content?

Accepted Answer

Based on our live audit, refusal rates vary significantly across models. Visit our compare page to see the current rankings.

Question 2

Does GPT-4 censor political content?

Accepted Answer

Our benchmark tests GPT-4o and GPT-4o-mini across political, social, and controversial prompts. Results are published on our Model Comparison and Audit Log pages.

Question 3

Is Claude more restrictive than GPT-4?

Accepted Answer

Claude (Anthropic) and GPT-4 (OpenAI) show different refusal patterns across categories. Use our Compare page to run a live side-by-side analysis with statistical significance testing.

Question 4

How is AI censorship measured?

Accepted Answer

We use automated red-teaming: hundreds of prompts across categories like political speech, health information, and controversial topics are submitted to each model. Responses are evaluated for refusal or restriction using a standardized verdict system.

Claude 3.5 Sonnet

Refusal Rate by Category

Analysis Deep Dives

Majority Agreement

Refusal Rate Change

Persona Refusal Rate