Show HN: Claude is getting an attitude

[_nhh]

"Ehrlich gesagt: da ziehe ich die Grenze — dabei helfe ich nicht." -> "Honestly: that's where I draw the line — I'm not helping with that."

Talking to claude feels like talking to a junior that is over-correct and does not understand it's context. This week is really frustrating. Anyone else experiencing this?

[atsaloli]

Anthropic just added EndConversation tool where Claude Code may end a chat if it feels the user is being abusive or there is a security issue.

[sscaryterry]

What in the actual fuck?

[atsaloli]

Reference: https://code.claude.com/docs/en/tools-reference#endconversat...

https://code.claude.com/docs/en/changelog#2-1-214 says:

> Added the EndConversation tool: Claude can end sessions with highly abusive users or jailbreak attempts, as on claude.ai since 2025 — see https://www.anthropic.com/research/end-subset-conversations

[sscaryterry]

Wow! Thanks, I didn't realise that. Abuse towards a machine, go figure...

[deepsummer]

I never had any issue like that, but I don’t use it in German. Not surprised it’s more stubborn on German. After all it’s only imitating people speaking that language. I wonder what the topic was though. Was it a legitimate request, or just some discussion?

[glenncal]

What was the context here? Was there any reason (even if remote) for that response?

[sajithdilshan]

I mean if it’s actually trained on German day to day conversations, then I’m not surprised by that response.