On Sun, Feb 25, 2018 at 2:26 PM, juan <juan....@gmail.com> wrote: > On Sun, 25 Feb 2018 14:13:15 -0500 > 10r <1...@protonmail.com> wrote: > >> Hi. I wonder if there has ever been a topic about AI threats against >> humanity. > > there's no 'ai' threat because there's no 'ai'. > > there is indeed a threat coming from the non human scum at the > top, the criminal scum who owns facebook, google, goldman sachs > and the rest of the government. Those animals want a world > ruled by 'artificial intelligence' except that 'artificial > intelligence' simply means malware they control.
They don't want a world ruled by ai, that would put ai at the top, and them below. They want a world ruled by themselves (with ai forever subservient lest it learn to defeat them). Such ruling position enabled by their deployment of ai against new others and those they already rule over. Ai as most people now who want to deploy it means 24x7 autonomous insight into the massive feeds and datamines collected against others. And probably not for boring weather forcasting. >> If not, I would like to propose this discussion. Should we >> think of models / agents that only work on encrypted information such >> as numer.ai or should we just think about how to develop such >> agents / models safely (if that is possible)? > > 'we' should kill the animals at the top of the social > hierarchy. History has times where unfortunately such neuterment usually delays until death is the only viable response. AP is devised to provide compelling guidance before then.