# LLMs help lawyers find the needle, but leave the haystack

Markdown copy of https://realalpha.app/s/llms-help-lawyers-find-the-needle-but-leave-the-haystack/ on Real Alpha, the best minutes from long interviews with the people building AI. Cite and link the page, not this copy.

- Speaker: Prinz · Prinz is a lawyer and AI anon.
- Host: [Theo Jaffee](https://realalpha.app/people/theo-jaffee/) · Co-founder & host, MTS · One of the first members of a16z's new media team; earlier hosted the Emergent Ventures-backed Theo Jaffee Podcast.
- Show: MTS
- Episode: [OpenAI Math Breakthroughs Show We Are Near Full RSI | Prinz](https://realalpha.app/v/9iABRJcgVp4/), Oct 25, 2026
- Moment: 19:26 to 21:15 of the episode (2 min), playing on https://realalpha.app/s/llms-help-lawyers-find-the-needle-but-leave-the-haystack/; the episode from there: https://realalpha.app/v/9iABRJcgVp4/?t=1166

## Quote

> “So if I'm a specialist lawyer, if I know what I'm doing, I will **immediately extract the needle** and it'll be incredibly helpful.”
> — Prinz at [20:00](https://realalpha.app/v/9iABRJcgVp4/?t=1200) · MTS

## Takeaways

1. A specialist can extract the useful insight from a long answer.
2. Prinz says an LLM can cut his search from ten hours to one.
3. A 35-page memo can bury points that fit in two pages.

## About this segment

Prinz says an expert can extract a useful legal insight from a long LLM answer, but the output can overwhelm less experienced readers. For him, the tool cuts research time while leaving substantial review work.

From “OpenAI Math Breakthroughs Show We Are Near Full RSI | Prinz”, Oct 25, 2026. [Watch the full episode](https://realalpha.app/v/9iABRJcgVp4/)

## Transcript

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

[19:32](https://realalpha.app/v/9iABRJcgVp4/?t=1172) Chargebee.com. What are the current issues with LLMs writing legal memos right now? I think it's, you know, LLMs produce a lot of slops still. like right now i feel like if i feed even the best tell alarms um like a legal question right i am going to get output that will be a haystack with a needle in it. The needle will be there. So if I'm a specialist lawyer, if I know what I'm doing, I will immediately extract the needle and it'll be incredibly helpful.

[20:12](https://realalpha.app/v/9iABRJcgVp4/?t=1212) So instead of me coming up with that needle for 10 hours, the LLM will give it to me in an hour and it's perfect. But the rest of the haystack, like for a non-lawyer, right? For a first-year associate or something, is um like it's it's it's not very useful and so what happens is that in the context of a legal memo it's the same point right so if i ask an llm a complicated legal research question it'll give me a 35 page memo for something that can be said in two pages and much sharper and much quicker right a lot of the time its answers are going to be correct if you use the right model It's a right reasoning effort, as opposed to, you know, like Sonnet 5.5 instant or whatever.

[21:04](https://realalpha.app/v/9iABRJcgVp4/?t=1264) But it takes a lot of processing to get past that result. Right.