**LLMs believe false statements even after explicit warnings that they're false*

**LLMs believe false statements even after explicit warnings that they're false**

> Fine-tuning tests show "bias ... toward confidently representing the claims as true." Do as I say, not as I say not LLMs believe false statements even after explicit warnings that they&\#8217;re false Fine-tuning tests show &\#8220;bias &\#8230; toward confidently representing the claims as true.&\#8221; Kyle Orland – May 28, 2026 5:29 pm | 76 This guy named Pinocchio really fed me some useful information in my training...

**来源信息**
- **来源**：Hacker News：AI 热帖
- **分类**：论文
- **发布时间**：2026-05-29 07:13（北京时间）
- **原文**：[打开原文](https://arstechnica.com/ai/2026/05/llms-believe-false-statements-even-after-explicit-warnings-that-theyre-false)