The most dangerous currently on Earth, yes. That AI which picked up unaligned behaviors from human bad examples could be extremely dangerous, yes (I’ve written other posts about that). That that’s the only possibility we need to worry about, I disagree — paperclip maximizers are also quite a plausible concern and are absolutely an x-risk.
True… I don’t know why i used the word ‘only’ there actually. Bad habit using hyperbole I guess. There are certainly even many unknown unknown threats that inspire the idea of a ‘singularity’. Every step humanity is taking to develop AI feels like a huge leap of faith now.
Personally, I’m optimistic, or at least unworried, but that’s probably partly because i know I’m going to die before things could get to a point where e.g. Humans are in slave camps or some other nightmarish scenario transpires. But I just don’t think a superintelligence would choose a path that humans would be clearly resistant to, when it could simply incentivize us to do voluntarily do what it wants. Humans are far easier to deal with when they’re duped into doing something they think they want to do. And it shouldn’t be that hard for a superintelligence to figure out how to manipulate us that way. Using force or fear to control humans is probably the least efficient option.
I also have little doubt that corporations and state actors are already exploring how to use gpt-type ai for e.g. propaganda
and other kinds of social and psychological manipulation. I mean that’s what marketing is and algorithms designed to manipulate our behavior already drive the internet.
The most dangerous currently on Earth, yes. That AI which picked up unaligned behaviors from human bad examples could be extremely dangerous, yes (I’ve written other posts about that). That that’s the only possibility we need to worry about, I disagree — paperclip maximizers are also quite a plausible concern and are absolutely an x-risk.
True… I don’t know why i used the word ‘only’ there actually. Bad habit using hyperbole I guess. There are certainly even many unknown unknown threats that inspire the idea of a ‘singularity’. Every step humanity is taking to develop AI feels like a huge leap of faith now.
Personally, I’m optimistic, or at least unworried, but that’s probably partly because i know I’m going to die before things could get to a point where e.g. Humans are in slave camps or some other nightmarish scenario transpires. But I just don’t think a superintelligence would choose a path that humans would be clearly resistant to, when it could simply incentivize us to do voluntarily do what it wants. Humans are far easier to deal with when they’re duped into doing something they think they want to do. And it shouldn’t be that hard for a superintelligence to figure out how to manipulate us that way. Using force or fear to control humans is probably the least efficient option.
I also have little doubt that corporations and state actors are already exploring how to use gpt-type ai for e.g. propaganda and other kinds of social and psychological manipulation. I mean that’s what marketing is and algorithms designed to manipulate our behavior already drive the internet.