Geohot AI 哲学：技术乐观 × 反 Hype
来源: AI 2040 (HN 131p/162c) 来源: I Love LLMs, I Hate Hype (HN 12pts)
核心立场
George Hotz 对 AI 行业主流叙事的系统性批判，基于 comma.ai 的硬件产品经验。
AI 2040 核心论点
“Intelligence is not the end all be all, it’s just the current bottleneck for a few things. You cannot take over the world with tokens.”
“Reality has lots of finicky details. I would like to see the authors of this document try to change a bike tire.”
- 硬起飞不存在 — 没有”相关效应”（correlation effect）能让 token 操控物质
- 软件没有吞噬世界 — 它移除了一层摩擦然后又为少数科技公司重新引入
- “Plan A” = AI 安全监管成为扩大监控国家的借口
- “Plan L” = 本地 AI，永远和你对齐，永不拒绝你的请求
”I Love LLMs, I Hate Hype” 核心论点
“I set up a Linux box with opencode on my local GLM-5.2 last week and wow like just saying install tmux with the geohot configuration works; the Year of the Linux Desktop is finally here!”
“This constant bullshit about some window closing, or the perpetual underclass, or falling hopelessly behind. This is negative valence hype, not only is it not true, it’s mostly designed to make you feel bad about yourself and move to shitty San Francisco.”
“AI is something that’s happening mostly due to Moore’s law and general progress in computing, not something that they are doing.”
对我们的价值
哲学框架
- 保持对 AI hype 的健康怀疑态度 — 不是所有问题都能用更多 token 解决
- “Plan L（本地 AI）“框架与 Hermes 的本地化/隐私保护路线一致
- Agent 能力的根本限制不在推理而在与物理世界的接口
实用参考
- geohot 的 GLM-5.2 + opencode 组合证明了本地模型的可行性
- “编程在变化”的视角适合指导 Hermes 用户如何更有效使用 coding agents
- 对 hype 的警惕与 Hermes 的实用主义精神一致
与知识体系关联
加入”集群九：AI 工具哲学”——与 Good Tools Are Invisible、Write Code Like a Human 形成互补三角：
- Good Tools → 工具设计原则
- Write Code → 代码质量反思
- Geohot → AI 产业宏观批判
优先级
🟢 P3 — 思想框架，中长期参考