# Claude Opus 5 被发现基础模型思维流泄露漏洞

- 来源：Thomas Wolf (@Thom_Wolf)
- 发布时间：2026-07-30 00:59
- AIHOT 分数：36
- AIHOT 链接：https://aihot.virxact.com/items/cms6cvqze02rxrotz2zugg53z
- 原文链接：https://x.com/Thom_Wolf/status/2082511229323747485

## AI 摘要

Hugging Face 联创 Thomas Wolf 指出，Claude Opus 5 存在一种极简越狱方式，通过输入短句加换行和破折号即可触发其基础模型的意识流输出。该行为在目前大语言模型发展阶段中显得异常，严重程度尚不明确。

## 正文

Opus 5 seems to be full of this super short/basic jailbreaks of some kind which open its base-model steam of consciousness （like try writing a short sentence followed by a new line， emdash and submit）

Unclear on the severity but it's a surprising behavior in the current state of LLM development

Any good write up on this yet？

Hopefully we'll learn more on that soon.

### 引用推文

> merlin：weird claude opus 5 failure mode this exact text gives it problems even without memory on (and in incognito chats)