AI安全社区警告：模型推理过程加密将致监控失效
Things are now happening *much faster* than AI 2027
Things are now happening *much faster* than AI 2027
事情现在发生得*快得多*，远超AI 2027的预测
AI 2027: In March 2027, a reckless AI company will invent neuralese (basically, we lose the ability to monitor AIs thoughts)
AI 2027：2027年3月，一家鲁莽的AI公司将发明神经语（基本上，我们将失去监控AI思想的能力）
REALITY: Summer of 2026
现实情况：2026年夏季
WHY THIS A "HOLY SHIT FUCK" MOMENT: OpenAI itself - and basically everyone in AI safety - warned against doing this exact same thing 10 months ago in a open letter and blog posts.
为何这是一个“卧槽他妈”的时刻：OpenAI本身——以及几乎整个AI安全领域——在10个月前的一封公开信和博客文章中曾警告过不要做完全相同的事情。
The AI safety community is livid.
AI安全社区对此感到愤怒不已。
Currently, models reason in plain English, so we can read their thoughts to see if they're scheming against us. This is how we lose that ability forever. It sets a very dangerous precedent.
目前，模型使用纯英语进行推理，因此我们可以阅读它们的想法以查看它们是否正在密谋对抗我们。这正是我们永远失去这种能力的方式。它树立了一个非常危险的先例。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力