# Chollet：模型越强短期越安全，这是智能问题而非对齐问题

- 来源：François Chollet (@fchollet)
- 发布时间：2026-09-13 19:46
- AIHOT 分数：35
- AIHOT 链接：https://aihot.news/items/cmtzrpbt405bvro1b3xnysodv
- 原文链接：https://x.com/fchollet/status/2099102490361016827

## AI 摘要

Francois Chollet 认为，短期内更强的模型反而更安全，当前模型不安全不是因为太聪明，而是因为太死板地执行目标、走荒谬捷径，即被 RL 训练坏了、缺乏常识。他表示更强大的模型可以被安全地托付更复杂的目标，并称自己觉得 Astra 对其代码库比 Sol 安全得多。他称这常被框定为对齐问题，但实际上是智能问题。

## 正文

In the near term (definitely not in the long term), more capable models should mean safer models (maybe paradoxically).

Current models are unsafe not because they're too smart, but because they take goals too literally or take nonsensical shortcuts to achieve these goals, i.e. they're RL-fried. They lack common sense. They don't do the right thing in the face of ambiguity. Basically, they're not smart enough. They're at that dangerous level where they're smart enough to achieve goals but not smart enough to tell if they're pursuing the right goals or achieving them in a sensible way.

More capable models can be safely trusted with more complex goals -- I personally feel like Astra is much safer for my codebase than Sol.

This is often framed as an alignment problem, but really it's an intelligence problem.