Commentary on X
@jasoncrawford Thanks for turning my rambling into something
3:04 PM · Jul 14, 2026"AI changes month to month. Institutions take years to build and decades to kill." Brian argues compellingly that the key step in AI safety is to learn more before we take hard-to-reverse actions like creating new global treaties or institutions.
"We must act now" is right in one dimension, I just don't think it's the dimension that gets featured in the NYT. Let's talk about a topic I've written on: policy can be thought through using real options theory 🧵
“If AI itself poses a large, irreversible harm, waiting is the risky move. … But institutions are irreversible too. And they can be harmful too” @BrianCAlbrecht
"We must act now" is right in one dimension, I just don't think it's the dimension that gets featured in the NYT. Let's talk about a topic I've written on: policy can be thought through using real options theory 🧵