Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Story

Anthropic tightens Claude policy against persistent abusive language

AI summary

科技新报 reports that Anthropic activated an updated usage policy for Claude on November 12 that prohibits persistent, unnecessary verbal abuse or cruel treatment. Normal complaints about incorrect answers, misunderstood instructions, or unsatisfactory results are exempt, while Claude may end sustained abusive conversations and severe violations may lead to account suspension. Searchable survey figures cited by the report say 33% of US chatbot users had insulted an LLM, with problem-resolution rates of 24% after abuse and 35% during rational interaction.

In-depth insight

The policy establishes clearer boundaries for Claude’s treatment of abusive prompts and could affect how users interact with the service, although Anthropic has not disclosed how often such abuse occurs.

1/3 美國用戶曾辱罵 AI,Anthropic 祭新規保護 Claude 免受「言語霸凌」

知名 AI 開發商 Anthropic 11 月 12 日正式啟用新版使用政策,特別針對旗下 AI 助理 Claude 增設一項明確的「防霸凌」條款:嚴禁使用者對模型進行持續且無必要的辱罵或殘酷對待。

正常抱怨免罰,但「惡意洗頻」恐遭停權

Anthropic 強調,這項新條款主要鎖定的是「反覆且沒有明顯目的」的極端辱罵行為。如果使用者是因為 AI 答錯問題、誤解指令,或是交出的成果不如預期而產生情緒性的正常抱怨,並不會受到處罰。

#Anthropic#Claude#Searchable

Original · 3 min readRead the original
Group chat

No comments yet. Start the conversation.