The weaponization has already started, and will just get worse. I wonder if we will need a sort of antidote ai, which generates fake content according to our values.

That is, it dramatically widens the impact of thoughtful, kind, caring, just, understanding, forgiving, etc. content, to counteract.