Transparency helps with normal "people problems." The opposite is true with bots, spammers, paid trolls, and other attempts to subvert the platform itself. It

Transparency helps with normal "people problems." The opposite is true with bots, spammers, paid trolls, and other attempts to subvert the platform itself. It's the same reason why open source projects try to fix security vulnerabilities privately before they disclose them publicly: they don't want bad actors to be able to take advantage of the information. Except with social platforms, they can't ever truly "fix" the problem, just mitigate it by trying to detect and remove offenders. If you provide transparency into how bad actors are removed, they'll just adapt to hide better.