Why would human language by the wrong interface when they're literally language models? Why would machine code be better when there is probably magnitude less of training material with machine code?

You can also test this yourself easily, fire up two agents, ask one to use PL meant for humans, and one to write straight up machine code (or assembly even), and see which results you like best.

> One could argue that LLMs learning programming languages made for humans (i.e. most of them) is using the wrong interface as well.

Then go ahead and make an argument. "Why not do X?" is not an argument, it's a suggestion.

because they are inherently text based as is code?

But they are abstractions made to cater to human weaknesses.

So you want LLMs to write a bunch of black box code that humans won’t be able to read and reason about easily? That will definitely end well.

Isn't that what LLMs are?

Not if you can review the code.