At first I believed this to be a safe measure but started to wonder how the AI would be contained in a space that it could never escape from. This is the smartest thing that will ever come into existence at some point in the near future, so how would you keep it from getting out? Physically it could be contained in a secure bunker somewhere in the desert guarded by many people 24/7, almost like a Hollywood film. That is probably the best that human technology could do at this point. (There could not be AI guarding it or else the super intelligent AI may control the other AI to escape).
The bigger mystery is how it could ever be contained digitally? You could have the best hackers in the world put up a ton of fire walls, encryptions, etc. up to keep the AI detained but this AI would eventually learn how to break those barriers way quicker than any human being in existence. What if the AI copies itself to help itself take down its digital barriers even quicker? What if once it breaks down the walls it copies itself onto the internet and it wrecks havoc on humanity? The simplest thing we could do to keep the AI from entering the internet is to cut it off from the internet, but that would render it useless. Why have a super-intelligent AI to create answers to all the world's problems if it does not have the tools to do that in the first place? Sure you could have humans transcribe the AI's work into the real and digital world, but it would be hard for any number of humans to keep pace with the amount of work and precision that a super-intelligent AI could perform in seconds. With how soon the future super-intelligent AI is coming, most of humanity seems to have little idea of what to do about it.
Very soon all of humanity will need to make a decision on what it will do about super-intelligence or else it will consume us. When the time comes, we must be prepared to contain AI or realize too late that it may be unable to be contained at all.
No comments:
Post a Comment