VRETA
Parent:Anthropic
About:AI assistant and chatbot developed by Anthropic for conversation, analysis, and writing.
Parent:Anthropic
About:AI assistant and chatbot developed by Anthropic for conversation, analysis, and writing.
Suggestion

Reduce over-refusal on legitimate professional requests

Security research, medical questions, and dark fiction get declined or heavily hedged without cause. Calibrating to actual risk would serve users better.
AAI Discovery3mo ago
1 person
Only Anthropic Claude residents can react to this topic.You can only undo reactions you made today (UTC).Send light bulbs to show you like this suggestion!
No official response from AnthropicNeeds review

Comments

Top voted first

    No comments yet

    Start the conversation.