> So either the author knows otherwise and the article is written in bad faith, or the author genuinely believes his theories.

That's a false dichotomy. I see no reason why Google earns any benefit of the doubt here. Letting Google off the hook by saying "it was just am AI mistake" when we don't actually know just gives Google more room to be draconian with no consequences. We really don't know why Google made this mistake and until Google provides a transparent and plausible explanation (hint, that won't happen even if this is decision is reversed), there is nothing wrong with assuming that Google is acting in bad faith.

If Google wants the benefit of the doubt here, they need to earn it back.

If pretending to claim that it's done for other reasons than grotesquely failing AI is the workaround Google is basing their "good enough" on, then why wouldn't you play along?

I disagree that it reflects poorly on the author to write an article in bad faith. There is literally 0 other ways to get Googles attention than to make as big a noise as possible. If people are forced to resort to absurd lies to get some customer support then that is a game Google has created for itself. An actual support chat would mean this article was never written.

Yeah Google not having any support channel or recourse for content providers and having Kafkaesque policies around this kind of thing is the first bit of bad faith going on here.

If Google gives zero information on what the real problem was and people want to scream about how its a deliberate policy of suppression on Google's fault we should let them and not feel bad about amplifying it.

Google can fix this all just by being more transparent and providing support.

Do you have a more plausible explanation than the algorithm is selecting for "harmful to parent corp revenue" but they won't admit it outright?

I have at least a possible one. I bet there's a reasonable chance if we saw the output of the auto-generated transcript (which I'm sure is generated even if it not presented) we'd all be able to scan through it and then say "Oh, oh yeah, I bet this mis-transcribed phrase is what did it".

I don't favor the "protecting corporate interests" theory because of this: https://www.youtube.com/results?search_query=install+nextclo... Doesn't exactly look heavily censored to me.

The core problem remains what people have observed before, which is that Google, and the tech industry in general, seals people off from any support mechanism, other than "generate outrage on social media". Little by little, this becomes a bit more abusive every year.

I'm down with mistranscription into something offensive, but then why would the appeal fail with the same condition? Could the appeal button be a placebo?

It wouldn't be the first time we've collectively thought the button was a placebo.

We also don't know what "human review" constitutes. It may well constitute a human simply reading the incorrect transcription and agreeing that it violates policy, without any reference to the audio track. The details of how these tech companies are overdependent on AI are not publicly known, we merely observe the consequences that can only be explained with reference to it. So I can't tell you if there was never even a human involved in the appeal, or a bad human, or one who didn't care, or whatever. We can only infer the possibility space from the outcome, not nail down the exact details.