4 ms·

I'll be keeping my body offline. But I think fusion with AGI is inevitable if we want to remain somewhat in control of our future. Otherwise we will be super ce

I'll be keeping my body offline. But I think fusion with AGI is inevitable if we want to remain somewhat in control of our future. Otherwise we will be super ceded by super intelligent AI and become their pets.

jakzurr 4y ago
Which actually might not be all that bad. Or we could go all conspiracy theory and suggest that we already are pets. ;)
- coenhyde 4y ago
  Hah, A lot of us already are pets, but at least the masters are still human.
jazzyjackson 4y ago
where do people get this idea that a flat plane of silicon can ever out-compete the chemical interconnectivity of fractally-folded grey matter (don't get me started on performance-per-watt) AI's greatest risk is people assuming it's superior while being little more than a charismatic magic 8-ball
- coenhyde 4y ago
  Who said anything about silicon? Maybe AGI will be built on our current bread of processors or maybe the next generation. The only way i see you can be correct is if today's exponential technological advancements come to a halt or reverse. I do not foresee a future where we do not eventually create an AGI superior to us. Maybe in 5 years time, maybe in a 100. But it is coming.
- pixl97 4y ago
  Because that fractally folder grey matter has only been generated via random walk and what survives to breed. It is also filled with massive stupidities that can end an individuals life in a moment. You're also thinking quite 2 dimensionally and limiting yourself to one medium. Carbon nanotube computing, chemical computing, even computing with DNA are all platforms that we can harness and have the potential to master.
trasz 4y ago
We wouldn’t even notice we’ve become pets of any kind of superintelligence; one can’t notice intelligence vastly superior to their own, same way your cat can’t even remotely grasp what the code on your screen is for.
- NateEag 4y ago
  Heh - this sounds kind of like theism. :)
  - trasz 4y ago
    Except that I'm not talking about some mythical "supreme beings", I'm talking about things that we've ourselves created. When you look at politics, you'll see that in many countries, from USA to Russia, "collective interest" matter more than those of humans; from corporations "sponsoring" regulations, to steadily lowering life quality and, at the same time, skyrocketing company income.
    - NateEag 4y ago
      Oh, okay. I personally wouldn't have called those things "intelligences" - more like clashes of unintended perverse incentives and ignored negative externalities. That said, I didn't think you were talking about a hypothetical deity - it's just what your metaphor made me think of.
      - trasz 4y ago
        >I personally wouldn't have called those things "intelligences" Why, though? By which criteria they are not intelligent beings?
        NateEag 4y ago
        Because in my experience these systems act dumber than most individual humans manage to. They function much more like a drunk, demented DNN optimizing for some unknown set of variables than like a human using its intelligence to solve problems. IMO. YMMV.
dekervin 4y ago
There was a span of 14 years between an implant allowi g control of an artficial hand, in 2005 , and an implant allowing speech synthesis from brain signals , in 2019 [0]. I am not sure the pace of innovation is accelerating drastically. (Edit: Sorry, I posted the wrong url in the previous version of my reply) [0] https://roboticsbiz.com/the-history-of-brain-computer-interfaces-bcis-timeline/ https://roboticsbiz.com/the-history-of-brain-computer-interf...

jakzurr 4y ago

Which actually might not be all that bad. Or we could go all conspiracy theory and suggest that we already are pets. ;)

coenhyde 4y ago
Hah, A lot of us already are pets, but at least the masters are still human.

jazzyjackson 4y ago

where do people get this idea that a flat plane of silicon can ever out-compete the chemical interconnectivity of fractally-folded grey matter (don't get me started on performance-per-watt) AI's greatest risk is people assuming it's superior while being little more than a charismatic magic 8-ball

coenhyde 4y ago
Who said anything about silicon? Maybe AGI will be built on our current bread of processors or maybe the next generation. The only way i see you can be correct is if today's exponential technological advancements come to a halt or reverse. I do not foresee a future where we do not eventually create an AGI superior to us. Maybe in 5 years time, maybe in a 100. But it is coming.
pixl97 4y ago
Because that fractally folder grey matter has only been generated via random walk and what survives to breed. It is also filled with massive stupidities that can end an individuals life in a moment. You're also thinking quite 2 dimensionally and limiting yourself to one medium. Carbon nanotube computing, chemical computing, even computing with DNA are all platforms that we can harness and have the potential to master.

trasz 4y ago

We wouldn’t even notice we’ve become pets of any kind of superintelligence; one can’t notice intelligence vastly superior to their own, same way your cat can’t even remotely grasp what the code on your screen is for.

NateEag 4y ago
Heh - this sounds kind of like theism. :)
- trasz 4y ago
  Except that I'm not talking about some mythical "supreme beings", I'm talking about things that we've ourselves created. When you look at politics, you'll see that in many countries, from USA to Russia, "collective interest" matter more than those of humans; from corporations "sponsoring" regulations, to steadily lowering life quality and, at the same time, skyrocketing company income.
  - NateEag 4y ago
    Oh, okay. I personally wouldn't have called those things "intelligences" - more like clashes of unintended perverse incentives and ignored negative externalities. That said, I didn't think you were talking about a hypothetical deity - it's just what your metaphor made me think of.
    - trasz 4y ago
      >I personally wouldn't have called those things "intelligences" Why, though? By which criteria they are not intelligent beings?
      - NateEag 4y ago
        Because in my experience these systems act dumber than most individual humans manage to. They function much more like a drunk, demented DNN optimizing for some unknown set of variables than like a human using its intelligence to solve problems. IMO. YMMV.

dekervin 4y ago

There was a span of 14 years between an implant allowi g control of an artficial hand, in 2005 , and an implant allowing speech synthesis from brain signals , in 2019 [0]. I am not sure the pace of innovation is accelerating drastically. (Edit: Sorry, I posted the wrong url in the previous version of my reply) [0] https://roboticsbiz.com/the-history-of-brain-computer-interfaces-bcis-timeline/ https://roboticsbiz.com/the-history-of-brain-computer-interf...