> I looked at a pre-AI project I did a couple of years ago: test to code LOC was 0.15:1. The projects I work on now are around 1.3:1.

Goodhart's law etc.

lazy low context response to a comment about lazy arguments.