Just as writers use grammar checkers and artists use digital brushes, developers increasingly rely on AI tools in their daily work. In this view, using AI assistance is no different than using any other resource available to a developer on the job.

That's clearly a broken analogy.

If you're smart enough to put this thing together -- then you're smart enough to see this fact as well.

What you've created here is a cheating tool, straight up.

The purpose of an interview is to find the best person to do the job, so that should be the criterium if we judge something as "cheating", not just tool use as such. Say you have a junior developer, fresh out of school, who's only just studied all the theory and done all LeetCode out there. They're going to do better on the traditional interview than a senior developer, whose more recent experience is more practical. But the senior developer is more qualified to do the job. Is it really cheating for the senior developer to use an AI tool to assist with the theory? Or is it merely levelling the playing field and making interviews fairer?

Your counterargument here seems to be: "Interviews suck because they don't measure the right things anyway. Therefore it's okay for candidates to use an AI tool during the interview process, even though employers explicitly forbid them from doing so."

I don't see any validity to this argument.

Except that it's a much more controversial technology than simply "digital brushes", at least as of now it is much more controversial. Could be widely accepted in the future of course

Only if you don’t plan on using AI when you’re working, and are misrepresenting your capabilities? Cheating is such a student concept.

Googling stuff and wrangling AI codegen soup doesn't scale past a certain point. I get "fake it till you make it" but there's a point to knowledge and mastery of a subject.

This is what the interviewer should be looking for. If you can fake it with an AI then they're not asking the right questions.