Longer writing lives on Substack. Shorter takes on X. Below are some ideas I keep coming back to.
Law school teaches you to spot risk. Every fact pattern is a potential liability, every contract clause a future dispute. This is useful—but it produces a profession that is structurally biased toward saying no. The lawyers who will matter most in the next decade are the ones who can look at a new technology and ask: what does this make possible?
Legal work is almost entirely language. Reading, writing, reasoning about text. This makes it one of the most natural applications of large language models—but also one of the hardest to get right. The stakes are high, the reasoning is subtle, and the tolerance for error is low. The companies that win will be the ones that understand the domain deeply enough to know where AI helps and where it doesn't.
I've never been able to sit still. As a kid, I built computers. In law school, I built an organization. After law school, I helped build a company. The common thread isn't technology or law—it's the desire to make something real. I think the world divides into people who analyze and people who build. I've always been on the building side, even when I was surrounded by analysts.