I’m sure that any flagged snippets will be submitted to a human for final review. They definitely won’t just auto-ban-hammer innocent people because the AI misinterpreted something they said!
Sigh.
and this filter will totally not only work for english and falsely flags and bans people using any other language, surely