Creatures from Primordial Silicon

This is a really quite impressively written article about the very beginning about applying machine learning to circuit design, from the late 90s.

Until I found the associated research paper, I at first wasn't sure if I was reading something real, or some elaborate, convincing creative sci-fi writing.

The original 1997 research paper can be read here: https://cse-robotics.engr.tamu.edu/dshell/cs625/Thompson96Ev...

Has anyone ever duplicated this, or at least tried to duplicate it?

I mean, it's quite common to have an alife experiment evolve an exploit if there's a bug in some part of the artificial ecosystem, so I wouldn't be surprised it it's easy to duplicate.