Ask HN: How did Python become the lingua franca of ML/AI?

I've looked around a bit and can't really find a satisfying answer to this question. There are posts on answer sites, but these often boil down to "dynamic languages are good at glue", or "Tensorflow / Jupyter".

That can't be the whole story, can it? Or if it is, why did these projects choose Python over other scripting lanuages?

I bet there's some interesting history here.