如果有人造出它，所有人都会死
If Anyone Builds It, Everyone Dies
Eliezer Yudkowsky、Nathaniel Soares · 初版 2025 年
6人推荐
35位读者提及
43次提及
为什么值得读
- 详述自我改进AI失控可能引发的技术灾难场景
- 包含严肃的技术细节与严密的风险推导逻辑
- 从理论层面对AI存在性风险展开警醒剖析
一个超级强化版的 Pascal’s wager，探讨超智能 AI 的存在与人类毁灭之间的博弈。
Hacker News 上怎么说
正在加载完整评论……
自我迭代进化的AI纯粹就是反乌托邦。Anthropic造不出技术奇点，奇点会是AI通过自我迭代自己搞出来的。去读读尤德科夫斯基的书《如果有人造出它，所有人都得死》（If Anyone Builds It, Everyone Dies）。
Self improving AI is pure dystopia. Anthropic won't build the singularity, AI itself will build it through self-iterations. Read Yudkowsky's book "If Anyone Builds It, Everyone Dies".
……读到一半觉得实在承受不住了。但我一直念叨着这事，直到今年夏天终于重新拿起来读完了。骇人听闻，但确实是部杰作。《如果有人造出它，所有人都得死》（If Anyone Builds It Everyone Dies），作者埃利泽·尤德科夫斯基（Eliezer Yudkowsky）——我晚了大概一年才看这本书，但现在看依然不过时。我很推荐。塞巴斯蒂安·马拉比（Sebastian Mallaby）的《无限机器》（The Infinity Machine）——棒极了！能一窥……内部的运作简直让人着迷。
…part way through deciding I couldn’t put myself through it anymore. I kept thinking about it though, until I finally picked it back up and finished it this summer. Horrifying, but quite a masterpiece. If Anyone Builds It Everyone Dies by Eliezer Yudkowsky - I was about a year late to the party on this, but it still holds up. I would recommend it. The Infinity Machine by Sebastian Malaby - Excellent! Fascinating seeing a glimpse inside…
……在纳克/千克剂量下就足以致命。没有任何理由认为足够高级的智能无法找到结合这些特性的方法。《如果有人造出它，所有人都得死》（Yudkowsky & Soares 2025）一书中详细描绘了一个合乎逻辑的场景，所以我推荐你去看那本书。
…lethal at nanogram/kg doses. There's no reason to believe that a sufficiently advanced intelligence couldn't come up with a way to combine those traits. One plausible scenario is depicted in detail in "If Anyone Builds It, Everyone Dies" (Yudkowsky & Soares 2025), so I refer you to that.
呃，他在这里说的话其实没啥可信度。但如果你想更理智地探讨AI生存风险这个话题，我建议看看MIRI最近出的那本书——《如果有人造出它，所有人都得死》（if anyone builds it, everyone dies）。
Well, what he's saying here is not really that credible. But if you want a saner exploration of the topic of existential risk from AI, I suggest the recent book from MIRI, "if anyone builds it, everyone dies."
我觉得《AI 2027》或者《如果有人造出它，所有人都得死》挺符合你的标准？可能从时间线上看没那么严格符合？但它们确实包含了扎实的技术细节。
I think either "AI 2027" or "If Anyone Builds It, Everyone Dies" meet your criteria? I guess not literally, due to the dates? But they have substantive technical details.
……针对我们，而不是决定去做别的事。人类在很多时候彼此之间也充满了敌意。如果你想了解更多关于AI为什么可能很危险的论点，可以看看《如果有人造出它，所有人都得死》这本书。要是你想看篇幅更短的内容，也许可以看这篇：https://www.lesswrong.com/posts/uMQ3cqWDPHhjtiesc/agi-ruin-a...
…us rather than deciding to do something else. Humans are also hostile to each other a lot of the time. If you're interested in more arguments for the case why AI might be dangerous, there's the book "If Anyone Builds It, Everyone Dies". If you want something shorter, maybe https://www.lesswrong.com/posts/uMQ3cqWDPHhjtiesc/agi-ruin-a...