Show HN: Claude is getting an attitude

#1

Post by _nhh » Tue, Jul 21, 2026, 6:54 AM UTC

"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?

Re: Show HN: Claude is getting an attitude

#2

Post by deepsummer » Tue, Jul 21, 2026, 10:11 AM UTC

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?

Re: Show HN: Claude is getting an attitude

#3

Post by atsaloli » Tue, Jul 21, 2026, 10:24 AM UTC

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.

Re: Show HN: Claude is getting an attitude

#4

Post by matchartier » Tue, Jul 21, 2026, 11:08 AM UTC

[dead]

Re: Show HN: Claude is getting an attitude

#5

Post by swn21 » Tue, Jul 21, 2026, 11:45 AM UTC

[dead]

Re: Show HN: Claude is getting an attitude

#6

Post by glenncal » Tue, Jul 21, 2026, 1:24 PM UTC

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

Re: Show HN: Claude is getting an attitude

#7

Post by sscaryterry » Tue, Jul 21, 2026, 1:28 PM UTC

atsaloli wrote: post #3

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.

What in the actual fuck?

Re: Show HN: Claude is getting an attitude

#8

Post by sajithdilshan » Tue, Jul 21, 2026, 1:36 PM UTC

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

Re: Show HN: Claude is getting an attitude

#9

Post by atsaloli » Tue, Jul 21, 2026, 9:36 PM UTC

sscaryterry wrote: post #7

atsaloli wrote: post #3

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.

What in the actual fuck?

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

Re: Show HN: Claude is getting an attitude

#10

Post by sscaryterry » Tue, Jul 21, 2026, 9:49 PM UTC

atsaloli wrote: post #9

sscaryterry wrote: post #7

Earlier quoted context omitted.

What in the actual fuck?

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

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