# 模型指令遵循与自主判断的矛盾

- 来源：Ethan Mollick (@emollick)
- 发布时间：2026-08-06 01:27
- AIHOT 分数：52
- AIHOT 链接：https://aihot.virxact.com/items/cmsgdgaei029bro5qer5coixq
- 原文链接：https://x.com/emollick/status/2085055055237914802

## AI 摘要

一个明显的奇怪矛盾：模型在遵循复杂指令方面越来越好，但同时也对指令中哪些部分最关注、哪些部分予以弱化，运用了更多“判断力”。

这对技能（skills）有重大影响——它们可能从命令变成建议。

## 正文

Noticeable weird contradiction: models are getting better at following complex instructions, but also using more "judgement" about which parts of the instructions they focus on most & which they de-emphasize.

Big implications for skills, which may become suggestions, not orders.