Anthropic says some Claude models can now end ‘harmful or abusive’ conversations

Back to Pulse

TechCrunchAug 16, 2025

Anthropic says some Claude models can now end ‘harmful or abusive’ conversations

Read the full articleAnthropic says some Claude models can now end ‘harmful or abusive’ conversations on TechCrunch

↗

What Happened

Anthropic says new capabilities allow its latest AI models to protect themselves by ending abusive conversations.

Our Take

We are tracking this story. Our take is coming soon.

What To Do

Check back for our analysis.

Cited By

TechCrunch Anthropic says some Claude models can now end ‘harmful or abusive’ conversations