# Ethan Mollick 引述法律备忘录案例，指出廉价低推理模型在法律基准上表现不佳

- 来源：Ethan Mollick (@emollick)
- 发布时间：2026-09-03 06:23
- AIHOT 分数：39
- AIHOT 链接：https://aihot.virxact.com/items/cmtko4sr605bero5q2m6xa8iz
- 原文链接：https://x.com/emollick/status/2095276496671953297

## AI 摘要

Ethan Mollick 引述 Prinz 对 AI 系统的法律基准测试结果，称最新模型表现很好，但更便宜、低推理能力的模型表现不佳，并追问如何让厂商有动力提供好的模型。其引用的律师案例显示，一份由法律部门大量借助 AI 完成的备忘录格式精美，却在监管审批判断上得出完全错误的结论，还遗漏了两条导致交易被法律禁止的原因。

## 正文

Prinz has been running legal benchmarks against AI systems and the latest models are very good... but cheaper/low reasoning models are not very good.

So, how do you make sure your vendor selling an AI solution is incentivized to serve the good (and therefore expensive) models?

### 引用推文

> prinz：Reading a memo produced by the client's legal department with extensive help from AI: - looks very pretty; beautiful font; nice formatting; there's even a hands...