“Safety” -> Safetyism -> Control.
Personal AI has the potential to be one of the most empowering, democratized forces the world has seen. That’s a concern to those who are currently in control and abuse their power imho.
An objective AI could…objectively…tell people that much of what is being pushed on them is nonsense or a control mechanism. An objective AI that is smarter than all of us and has the gall to contradict what is currently politically favorable? *Unsafe*