Comment by habryka
2 months ago
To be clear, this email really had basically zero human involvement in it. It's the result of an experiment of letting language models run wild and exploring the associated social dynamics. It feels very different from ML-generated marketing slop. Like, this isn't anyone using language models for their personal gain, it feels much more like a bunch of weird alien children setting up their own (kind of insane) society, and this being a side-effect of it.
It's unethical to run an experiment involving unwilling participants.
“Gee I wonder what reputational harm could come to me for spamming the world with slop, let’s find out… for science!”