Benchmark RadarAI BENCHMARK PROFILE
AuAu
AuAu combines psychometric instruments, vignettes, and realistic prompts to assess authoritarian tendencies in LLM responses, measuring sub-concepts like aggression, submission, and conventionalism.
- Released
- 2026-06-15
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Provides a structured approach to auditing LLM authoritarian alignment, highlighting variations across models and the impact of system prompts on authoritarian output.
Motivation
The worldwide rise of authoritarianism and the growing role of Large Language Models (LLMs) in users' everyday lives raise the question of whether specific models exhibit or promote authoritarian attitudes.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.