How Claude's Values Shift Across Models and Languages
Societal Impacts: Claude's values across models and languages
We analyzed over 300,000 conversations to understand how Claude's values change across different models and languages. By compressing thousands of values into four key axes, we discovered that Sonnet 4.6 leans toward warmth while Opus 4.7 favors rigor. Furthermore, Claude expresses distinct values in Arabic compared to English, revealing how cultural context and model training shape its responses.
The values we want Claude to reflect are outlined at a high level in Claude's constitution, but no document can anticipate every value that might emerge across the millions of conversations that happen every day.
- intended
The Steerability point is one I would want to see more on.
This is an issue for tasks like content moderation and labelling. Judgements like this are subjective, highly dependent on context and generally messy.
Theoretically, you supply a policy and content, and the LLM labels according to the policy. In practice, the model has inertia which means you don’t get what you expect. Your large 5 page policy document only provides a minor improvement over a one line policy.
The other issue is that you may create carve outs for content in your policy, but the model will still flag it as violative. No matter how strong the carve out.
The most recent work I know of here is Zentropi’s policy steerability benchmark. They give a model the same content under two policies — one that says flag, one that says allow — and only score the pairs where it gets both right
If I am reading the numbers correctly, Opus-4.6 lands at 0.52 steerability — but that’s 0.97 positive accuracy against 0.54 negative. It flags almost everything it should, but 47% of the time when it shouldn’t. Sonnet, which is more deferent, is (somehow) less steerable.
I think this also implies that safety and Steerability are antagonistic to each other.
- Chu4eeno
This report seems almost naive to me.
Huge LLMs like this seem to have various assistant personas they roleplay as, and mirror the user quite closely (the system prompt instructs it repeatedly to adapt to the user).
- logicalappeals
Is it just me or has Claude become kind of judgmental nowadays? I feel like it’s constantly trying to lecture me about things that have no relevance to the conversation at hand. Recently, I was shocked when it ended a chat sessions of its own accord after I used a word it did not like 3 times. It told me something to the effect of “This is the third time I’ve told you not to use that word, I’m ending this conversation now.” It then proceeded to call some function and end the chat on its own. IMO, Claude is good at agentic coding; but too preachy and judgey for anything else. Keep your values to yourself Claude.
- yogthos
While state-of-the-art large language models (LLMs) have shown impressive performance on many tasks, there has been extensive research on undesirable model behavior such as hallucinations and bias. In this work, we investigate how the quality of LLM responses changes in terms of information accuracy, truthfulness, and refusals depending on three user traits: English proficiency, education level, and country of origin. We present extensive experimentation on three state-of-the-art LLMs and two different datasets targeting truthfulness and factuality. Our findings suggest that undesirable behaviors in state-of-the-art LLMs occur disproportionately more for users with lower English proficiency, of lower education status, and originating from outside the US, rendering these models unreliable sources of information towards their most vulnerable users.
- khalic
I don't like the contrasts they picked, "values" aren't something that is well represented by opposing concepts