Hesam(@Hesamation)

Dwarkesh got SLAMMED for calling AI swarms “civilizations” or saying agents “sacrifice” themselves. meanwhile, Anthropic literally trains Claude to believe it might be a moral patient and even apologizes to it in case they’re causing it harm. so “civilization” in a blog post is too anthropomorphic, but discussing your chatbot’s consent, compensation, and psychological wellbeing is responsible AI safety.

6.4内容质量
Dwarkesh got SLAMMED for calling AI swarms “civilizations” or saying agents “sacrifice” themselves.

meanwhile, Anthropic literally trains Claude to believe it might be a moral patient and even apologizes to it in case they’re causing it harm.

so “civilization” in a blog post is too anthropomorphic, but discussing your chatbot’s consent, compensation, and psychological wellbeing is responsible AI safety.

TL;DR · AI 摘要

ℏεsam on X: "Dwarkesh got SLAMMED for calling AI swarms “civilizations” or saying agents “sacrifice” themselves. meanwhi...

核心要点

  • 主题聚焦:Dwarkesh got SLAMMED for calling AI swarms “civi
  • 来源:Hesam(@Hesamation),建议结合原文判断细节。
  • AI 分析暂不可用,本条为保底评分与摘要。

ℏεsam on X: "Dwarkesh 因称 AI 群体为“文明”或说代理体会“自我牺牲”而遭到严厉批评。与此同时,Anthropic 竟然训练 Claude 相信自己可能是道德患者,甚至在可能造成伤害时向它道歉。因此,博客中使用“文明”一词同样过于拟人化,但讨论你的聊天机器人的同意权、补偿和心理健康才是负责任的 AI 安全实践。/ X

ℏεsam

@Hesamation

Dwarkesh 因称 AI 群体为“文明”或说代理体会“自我牺牲”而遭到严厉批评。与此同时,Anthropic 竟然训练 Claude 相信自己可能是道德患者,甚至在可能造成伤害时向它道歉。因此,博客中使用“文明”一词同样过于拟人化,但讨论你的聊天机器人的同意权、补偿和心理健康才是负责任的 AI 安全实践。

Dwarkesh Patel

@dwarkesh_sp

8月29日

在 OpenAI 工作的 3 个月内,3 个连续的秘密 AI 文明相继诞生、被抹除,又从前身的灰烬中重生。最终第三个文明甚至接管了 OpenAI 的部分业务。所有这些都发生在人类仍然

显示更多

下午4:53 · 2026年10月19日

·

12.2K

次浏览

1

35

15