Oh no, maybe the V-Model was right all the time? And right sizing increments with control stops after them. No wonder these matrix multiplications start to behave like humans, that is what we wanted them to do.

So basically you’re saying LLMs are helping us be better humans?