# 斯坦福研究：Gemini 2.5 Pro 法律问答胜率超人类教授

- 来源：Ethan Mollick (@emollick)
- 发布时间：2026-06-03 02:24
- AIHOT 分数：47
- AIHOT 链接：https://aihot.virxact.com/items/cmpwzdg2b027fsl79rzyycgd0
- 原文链接：https://x.com/emollick/status/2061876620638486584

## AI 摘要

法学教授们写下了他们在办公时间被学生问到的问题。Gemini 2.5 和人类分别作答，然后其他法学教授在不知道答案作者的情况下对结果进行了评判：
- Gemini 的胜率为 75%，击败了教授们
- Gemini 的答案被评为比人类的答案危害更小
- 更新的模型表现甚至更好

## 正文

Law professors wrote questions they were asked during office hours. Gemini 2.5 &amp； humans answered them then other law professors blindly judged the results：
-Gemini had a 75% win rate vs. professors
-Gemini's answers were rated LESS harmful than humans
-Newer models do even better

### 引用推文

> Andrew Curran：In a new Stanford study, law professors by far preferred Gemini 2.5 Pro's responses over those written by their peers when they were unaware of who wrote the an...