# Anthropic 新研究：Claude 能否自主对齐其他 AI

- 来源：Anthropic (@AnthropicAI)
- 发布时间：2026-08-29 01:13
- AIHOT 分数：51
- AIHOT 链接：https://aihot.virxact.com/items/cmtd86t1701dhro666couferh
- 原文链接：https://x.com/AnthropicAI/status/2093386528668172373

## AI 摘要

新研究员研究：Claude 能否自主对齐其他 AI？

我们给了 Claude 48 小时和 1 块 GPU，让它改进小型模型的对齐。它自行研究并提出方法，然后自主训练和测试模型。效果出乎意料地好。https://www.anthropic.com/research/automated-researchers-mitigate-alignment-failures

## 正文

New Fellows Research: Can Claude autonomously align other AIs?

We gave Claude 48 hours and 1 GPU to improve the alignment of small models. It researched and proposed methods, then trained and tested the models on its own. It worked surprisingly well.
https://www.anthropic.com/research/automated-researchers-mitigate-alignment-failures