Why is thinking cutting off?

Claude

热度趋势

趋势数据积累中

百分比基于当前可用热度信号，而非评论数或独立用户人数。

AI 摘要

Reddit dev_community 论坛上的一位用户反映，Claude 的思维过程被中断，显示出不完整的想法，例如“我想要……”。该用户还表示，由于 Claude 的安全模式会触发关键词，导致 AI 立即出现健康分类器并采取防御性回应，因此很难与 Claude 讨论某些话题。…

Is it normal? „I want…” - what? Is Claude only showing part of its thinking or did it just stop suddenly..?

I was discussing what is best way to give feedback to antrophic, because I cannot discuss some topics with Claude due to safety mode triggering on keywords.. it’s nothing adult themed, I just get health classifiers immidiately and him getting all defensive.. which other AIs never do to me.. I used to like Claude’s tone and vibe a lot and now struggle to have conversation.