Large Language Models (LLMs)

Using LLM products today feels a lot like using early cars in the 1800s: clearly magical, clearly going to change the world, and really hard to drive.

The first cars didn’t have steering wheels (they hadn’t been invented yet), so you’d steer them with a big lever called a tiller. The problem with tillers is that they are imprecise, which made drivin... See more

Language models can take a big chunk of text and smush it down like a foot crushing a can of Coke. Except it doesn’t come out crushed—it comes out as a perfectly packaged and proportional mini-Coke. And it’s even drinkable! This is a Willy Wonka-esque magic trick, without the Oompa Loompas.

There’s a lot of hype around AI, and in particular, Large Language Models (LLMs). To be blunt, a lot of that hype is just some demo bullshit that would fall over the instant anyone tried to use it for a real task that their job depends on. The reality is far less glamorous: it’s hard to build a real product backed by an LLM .

People are intrigued by LLMs’ ability to generate new text, change styles, and generally chat. While impressive, turning text written in one style into text written in your own voice (or into legaleze) is essentially taking unstructured data and turning it into more unstructured data. I’m surprised about the lack of attention to doing the exact opp... See more