What instantly came to mind for me was that this is yet another step forward in turn-developers-lives-into-dystopias. It wasn't bad enough that the algorithms would flag and ban-hammer people with no ability to reach a human, but now we have to use "AI" to make those decisions. Hopefully not the same AI that routinely fails to accurately summarize news articles... And don't worry, once the AI decides the person is evil, it proceeds to nuke everything associated with them as well so that way their personal photos in G Photos and personal gmail account etc will also disappear, and ideally they won't even be able to use their Android phone.

I have little sympathy for (truly) bad actors, but it seems we've completely abandoned Blackstone's ratio in favor of returning to a world where the nobility can execute a peasant without a trial. Now the human doesn't even have to hear the peasant's explanation or petition!

I would never do a startup nowadays that relied on shipping an app to the Play Store (or the Apple store for that matter, but that's a different cause which is off-topic atm). In fact, I've pivoted my current org to focus first on developing a great web app experience, and the native apps second. Putting your livelihood or company on the mercy of these app store people is a terrible and increasingly inexcusable risk.