He misses the point that AI is purposeless and there is always a human behind it’s use. Like in the campaign example: why would AÍ get intimate with someone and change their views? Because a human asked to. And what if the change in view is for better?

Reply to this note

Please Login to reply.

Discussion

There has always been competing uses of technology (for good and bad) and it’s the success rate of the good uses, over long term, that protects us from Armagedom, not regulatory agencies. Definitely not!