I have been responsible a bit of content moderation myself, and interacted with mods on various forums. It's possible that this is responsible for some of my views, summarized here. - The "secret" aspect of all this can be largely ignored. Twitter is under no obligation to tell its users exactly how their Tweets are moderated. I am sure most moderation tools are secret, including whatever Twitter (or for that matter,…
1) We should censor for how things are said, instead of what is said. When censorship is carried out secretly it will inevitably trend towards abuse. One of the examples given in these leaks is that @DrJBhattacharya [1] was secretly put on a trends blacklist because of what he was saying. He's a well spoken doctor and Stanford epidemiologist, but he publicly spoke against the political measures being taken against COVID, such as lockdowns. So he was blacklisted. That's just so incredibly wrong, and enabled only by secrecy.
2) Again with the goal of adjusting how things are said, instead of what is said: If you think somebody is an idiot, saying that accomplishes absolutely nothing besides sending discourse to the level of a dumpster dive at a cheap seafood joint on a hot summer day. By contrast, engaging in good faith and simply discussing the topic not only keeps the conversation healthy, but might even actually change some minds. Secretly censoring the shitposter does not really encourage him to engage in a more productive fashion, because he may not even realize he's behaving a way that's not really considered acceptable, let alone attempt to correct it.
---
Literal spam is a different topic, and I think conflating the two is disingenuous. Doing things like blacklisting doctors because of what they say has little to do with banning somebody trying to impersonate other accounts and spamming 'give me 1 and I'll give you 2' crypto scams.
[1] - https://twitter.com/DrJBhattacharya
[2] - https://www.nytimes.com/2022/01/04/briefing/american-childre...