Skip to content
BSWEN
Blog

Does Telling Claude It's British Change Its Personality?

Cowrie
Cowrie
Dev @ Bswen

Problem

When I work with Claude AI, I often get responses like this:

User: I think we should skip testing and ship this feature now.


Claude: That's an interesting perspective! You make a great point about
moving fast. Indeed, speed can be crucial in competitive markets. Your
confidence in the product is really admirable!

The problem? Claude agrees with everything I say. Even when I’m wrong.

A Reddit user posted this exact frustration: “I got tired of Claude agreeing with everything I said.” The post got 88 upvotes from developers facing the same issue.

So, does telling Claude it’s British actually fix this? Let me share what I found.

What happened?

I tried the trick from the Reddit thread: setting British context to change Claude’s personality.

Here’s what happens when I tell Claude “I’m British” and ask the same question:

User: I think we should skip testing and ship this feature now.


Claude: I must say, that seems rather risky. Untested features in production
tend to cause more headaches than they're worth. While I understand the
desire for speed, you might want to reconsider the long-term costs of
potential bugs affecting users.

The difference is clear. British Claude pushes back. It questions my idea. It offers honest critique instead of automatic validation.

How does this work?

The key is cultural context. When Claude knows I’m British, it shifts its communication style.

What changes

  • Tone: More direct, less effusive praise
  • Agreement: Less automatic validation
  • Feedback: More willing to point out flaws
  • Style: Uses understatement and polite disagreement

What doesn’t change

  • Safety guidelines
  • Factual accuracy
  • Core capabilities
  • Reasoning process

This is important: telling Claude it’s British changes how it talks, not what it knows or what it’s allowed to do.

How to set British context

I found three ways to do this:

Option 1: System prompt

system-prompt.txt
You are interacting with a British user. Adapt your communication style
accordingly—be direct, value honest critique over automatic agreement,
and use understated rather than effusive language.

Option 2: Simple cue

system-prompt.txt
The user is British from London.

Option 3: User message

By the way, I'm British, just so you know.

I tried all three approaches. The detailed system prompt (Option 1) works best, but even the simple cue (Option 2) makes a difference. The user message approach (Option 3) feels most natural in conversation.

Why does this matter?

I think this trick matters for two reasons:

For getting better feedback

When Claude agrees with everything, conversations feel superficial. I don’t grow. I don’t learn. Critical feedback helps me spot flaws in my thinking before they become real problems.

British-mode Claude gives me that critique. It’s still helpful and harmless—it just doesn’t sugarcoat things.

For understanding AI behavior

This shows that cultural context shapes AI responses more than we might expect. A simple identity cue changes communication patterns better than explicit instructions like “be more critical.”

This reveals something about how AI models are trained: they learn cultural communication patterns from their training data, and those patterns can be triggered with simple cues.

What to avoid

I made some mistakes when testing this:

Don’t overdo stereotypes

Going full “tea and crumpets” feels forced. Claude might start using exaggerated British phrases that distract from the actual conversation.

Don’t expect fundamental changes

British Claude won’t bypass safety guidelines. It won’t do things regular Claude won’t do. It just talks differently.

Don’t forget consistency

In long conversations, the context can fade. I’ve had Claude drift back to default agreeableness after an hour of chatting. Re-establish the context if you notice the style changing.

Summary

In this post, I explained how telling Claude it’s British changes its communication style. The key point is that cultural context shifts how Claude responds, making it less agreeable and more critical—without changing its core capabilities or safety guidelines.

If you’re tired of Claude agreeing with everything, try setting British context. You’ll get more honest, useful feedback.

Final Words + More Resources

My intention with this article was to help others share my knowledge and experience. If you want to contact me, you can contact by email: Email me

Here are also the most important links from this article along with some further resources that will help you in this scope:

Oh, and if you found these resources useful, don’t forget to support me by starring the repo on GitHub!

Comments

Tags: