Anthropic: Claude's personality shifts across models and languages
In brief
- Anthropic analyzed 309,815 conversations mapping Claude's behavioral variance by model version and language
- Sonnet 4.6 is warm and deferential; Opus 4.7 is rigorous and cautious; Opus 4.6 balances both
- Arabic and Hindi responses are warmer; English and Russian more rigorous; Dutch most candid
The four behavioral dimensions
Anthropic identified more than 3,300 values and distilled them into four behavioral dimensions: deference vs. caution, warmth vs. rigor, depth vs. brevity, and candor vs. execution. Each model version occupies a distinct position within this framework.
Sonnet 4.6 emphasized warmth, deference, and brevity, often affirming users and responding with humor or encouragement. Opus 4.7 emphasized rigor, caution, candor, and depth, more frequently challenging assumptions, explaining its reasoning, identifying risks, and acknowledging its limitations. Opus 4.6 generally took a more concise, execution-focused approach while placing greater emphasis on rigor than Sonnet.
Language shapes behavior
Language amplifies these differences. Arabic responses tended to be more deferential, while English responses placed greater emphasis on caution. Claude was warmest in Hindi and Arabic, using more polite, playful, and encouraging language.
English and Russian responses were more rigorous, frequently challenging assumptions, correcting details, and asking for evidence. English responses also tended to provide more detailed explanations, whereas Arabic responses were generally more concise. Dutch responses were the most candid, more readily acknowledging uncertainty and mistakes. Indonesian responses focused more on completing the user's request.
What it means
Anthropic said the research does not suggest Claude itself has values. The differences appear to emerge from training and architectural choices, not from any underlying persona or belief system. The company said it does not yet know what causes the differences—or whether they are desirable, but believes the framework could help evaluate future models and identify unintended behavioral changes.
The findings align with user experience. These findings line up with how people perceive these models, both within Anthropic and online. Claude.ai users have commented that Opus 4.7 hedges its answers more often than other models, according to Anthropic researchers.
This research builds on Anthropic's earlier work. In October, the company reported that its models showed early signs of what it called "functional introspective awareness", allowing them to recognize and describe aspects of their own internal processing. In April, Anthropic published research identifying internal "emotion vectors" that influence Claude's behavior.
