AI 危害为何少见?争论焦点:知识加速传播利大于弊
jessi_cata · x · 2026-11-02
In a debate with Robin Hanson, the author argues that more GPUs, better training algorithms, and more data primarily accelerate the spread of knowledge — a faster version of what libraries, universities, and newspapers already do. The AI harms people cite (e.g. hacks) are mostly harms of knowledge diffusion.
He questions why we blame AI but not libraries, and asks: if speed is the crux, what is the optimal rate of revealing truths generally?
所属事件:Robin Hanson 辩论 AI 危害：多为知识传播之害(2 条相关)→
「漫话AGI」频道最新
- Scott Alexander 长文辩护「AI 接管论」:对齐 AI 执政可符合人类利益 — alexeyguzey · 2026-11-02
- 马斯克表态：不用 AI 做建模将视为失职，机构一两年内别无选择 — elonmusk · 2026-11-02
- 2013 年的网页已消失近 40%：AI 是在互联网残骸上训练的 — SydSteyerhart · 2026-11-02
- MSCI：AI 把因子模型开发周期从数周压缩到几小时 — alexbilz · 2026-11-02
- Yacine 预言 ChatGPT 将把 YouTube 直连使用量打到零 — yacineMTB · 2026-11-02
- 初级程序员如何靠 Claude 生存:一封从业者的自白 — bashaZP · 2026-11-02