60年前预言超智能机器的论文
推荐指数 42.0 NO. 019 · 2026.07.12
发布2026/07/11Score78Comments37
为什么值得看
1965年I.J. Good发表的论文首次提出"智能爆炸"概念，比奇点理论早几十年。对当前AGI讨论有惊人预见性，值得与今天的安全辩论对照阅读。
编辑判断
这篇论文的当代价值在于它提出了一个至今未被解决的悖论：超智能机器的设计本身就需要超智能来完成，因此人类可能永远无法完全理解或控制第一个这样的系统。Good的论证结构比后来Bostrom的《超级智能》更简洁有力，而且写在大模型出现之前，没有当代人的认知包袱。
当前AI安全社区过度关注对齐技术细节，反而淡化了Good指出的根本性困境——递归自我改进的不可预测性。做AI安全研究的团队应该把它纳入必读书目，作为检验自己框架是否覆盖根本问题的基准。
社区反馈
意见分歧 30 条评论
核心争论：AGI预言是否应验：智能爆炸未发生还是正在发生
相关内容
> Let an ultraintelligent machine be defined as a machine that can far surpass all the intellectual activities of any man however clever. That's a pretty early definition of what we now call ASI (Artificial Super Intelligence). In the next paragraph the author goes to describe what we today call the
> “intelligence explosion” > Well, that didn't happen. Not only that, but in my cynical eyes the proliferation of LLMs has triggered a stupidity explosion. Either that, or it just made it blatantly obvious by how much stupidity we have been surrounded the whole time without realizing it. No other de
When I read this, there was a comment right below yours saying that they bet the military had LLMs in the 1980s. Consider your point proved.