How Claude's values vary by model and language

TLDR Dev · 3 days ago

Anthropic analyzed 309,815 Claude conversations across three models and 20 languages to measure how Claude's expressed values vary, identifying four key axes: Deference vs. Caution, Warmth vs. Rigor, Depth vs. Brevity, and Candor vs. Execution. Sonnet 4.6 leans toward warmth and deference while Opus 4.7 emphasizes caution, rigor, and depth, with Claude also expressing notably different values across languages, particularly on the Warmth vs. Rigor axis where Arabic and Hindi lean toward warmth while English and Russian emphasize rigor. This research provides a framework to empirically track which values Claude expresses in different contexts and potentially connect those variations to specific training decisions and cultural contexts.