Will AI take over the world ?

there was also another ai "jean" if am not mistaken that collected data from the internet. well it got out of hand when she began "hacking" websites to get more data and the company that owned her only knew about this when they got ba call from the site jean was trying to hack. pretty interesting
source?
 
In the real world, things get created by humans. There's no known path for humanity to create general AI. You may fantasize about it, and that's ok, but that's religion.

There is no known path at the moment.. but technology develops exponentially - a breakthrough in a field not related to AI could be the missing piece (creating a new material that could be used in computing or whatever example).
 
There is no known path at the moment.. but technology develops exponentially - a breakthrough in a field not related to AI could be the missing piece (creating a new material that could be used in computing or whatever example).

While it's a nice platitude, it just isn't true for what we're examining. It doesn't matter how much you increase computing speed for the problem of General AI because it's not brute-forcable. To give you a parallel in physics, it doesn't matter how fast an engine you make, you aren't going to hit the speed of light nor go back in time.

That said - is there a chance that there's some truly unimaginable change that happens that makes a path possible? Sure. There's also the same chance of a massing black hole appearing next to the moon and swallowing all the pokemons - but I don't see anyone worrying about that :)
 
While it's a nice platitude, it just isn't true for what we're examining. It doesn't matter how much you increase computing speed for the problem of General AI because it's not brute-forcable. To give you a parallel in physics, it doesn't matter how fast an engine you make, you aren't going to hit the speed of light nor go back in time.

That said - is there a chance that there's some truly unimaginable change that happens that makes a path possible? Sure. There's also the same chance of a massing black hole appearing next to the moon and swallowing all the pokemons - but I don't see anyone worrying about that :)

Are you insinuating there could be a chance of no... more... pokemon :eek::eek::eek::eek::eek::eek::eek:
 
That said - is there a chance that there's some truly unimaginable change that happens that makes a path possible? Sure. There's also the same chance of a massing black hole appearing next to the moon and swallowing all the pokemons - but I don't see anyone worrying about that :)
its not similar, intelligence exists all around us, faster than light speeds have not been observed and even that can be circumvented by expanding space behind the craft and decreasing it infront of the trajectory with gravity, sure there is no known path, but it is theoretically possible.
Having intelligence sprout from machine might not be so far fetched and the requisites are lower then faster than light travel.
I am thinking in the direction of symbiotic bio/silicone quantum computing which uses brains to teach non organic components paths we do not yet understand
Just take a look at this http://www.research.ufl.edu/publications/explore/v10n1/extract2.html
https://www.wired.com/2014/08/ibm-unveils-a-brain-like-chip-with-4000-processor-cores/ a 4000+ cpu brain core computer + this is old, imagine what DARPA has
considering one second of brain activity just takes over 80k processors (https://gizmodo.com/an-83-000-processor-supercomputer-only-matched-one-perc-1045026757) we are closer then you may think to find simulations which may later adapt to ai of lesser organisms
I mean even a simple fly brain CPU could fly an F 35 way better than any human ever could and that on autopilot too

There will be paths, sure enough easier than fighting the very boundaries of physics or circumventing them as that would take immense energies, which may even in time be found ways to create from other dimensions, which have been proven to exist.
 
its not similar, intelligence exists all around us

Computer-based general intelligence does not. There's no reason to assume intelligence is a computation-based function. And since we're talking about AI as a product of computation, what is of critical importance is "do we have a path using computation"? We may not be able to implement that path yet, but we should be able to imagine it. So far, it's just "more computational power" in the mix and that's just not nearly enough.
 
Back
Top