E.g. with an existing technology known as firearms, it's not socially 'OK' or legal to just go up and randomly shoot someone with such technology, but technically you could do it, if you really wanted. There's consequences to it though.
So we need consequences for violating people's privacy or using their information in a way that is abusive or unfair to their human rights.
Radical transparency of disclosure such as what insurance companies know about you that they feed into their business will have to be legally mandated. (How about an open-source algorithm too?) We will need new frameworks by which data collected on you and used against you MUST be disclosed, because otherwise it could be total abuse and fake. We know China is already in that territory.
The world of deepfakes is going to make 'deniability' a major opposing force of surveillance as well.
We have an interesting future ahead of us.