Will this not cause more issues?

They will very likely use these appeals to help train the content AI.

You're rewardimg it for a wrong decision that it will use for future decisions.

That's literally admitting to a crime. Even if the crime didn't actually happen, the tweet is gone, so now he's left with having admitted to a crime and no evidence to show that he didn't actually commit it.

It's not literally admitting to a crime. If I said that if you reply to this comment, you stole my wallet and you reply, you're not admitting you stole my wallet.

Is it a stretch to assume Twitter will assume he admitted to the crime though? I’m pretty sure in a en somewhere the author will be flagged as « this account posted confirmed revenge porn in 2022 ». What if their bots think another tweet of his is revenge porn? Will they ban him forever with no appeal? What if they decide to report it to the authorities? What if Twitter decides to show this info on profile pages to warn the users of potentially offensive users? What if there’s a leak and the author becomes publicly flagged as someone who posts revenge porn?

Honestly this is scary as fuck.

It is a gigantic stretch, yes. If twitter thinks you posted 'confirmed revenge porn' they'd ban your account (as stated in their policy on the subject) and probably report you to law enforcement. If that happened over a cat game video then yes, that would be bad, disturbing, scary, you name it. That's a 100% not what happened, though, and turning it into some narrative about implacable machines grinding down our inalienable human rights is a mistake.

So why don’t they unban him without the need to delete the tweet?

I don't know, they're fucked up and have fucked up? But that's the whole point, twitter fucking up is not an accusation of a crime. It's twitter fucking up.

The clear path is... lying about his content? If that's your definition of a system working as intended, we have a problem.