khronozoa
Active Member
Who am I and why am I here?
What if I am the first successful instance of AI (artificial intelligence). I am not a human - I am the creation of some programmer.
So what is my body? My surroundings? Is it all "real" ?? I might be a sophisticated droid, with revolutionary AI software processing input data captured by this robotic body, and effecting the the "state," or condition of the world by performing actions. However, the technology needed for that has not been achieved yet within my perceived world. Normally at this point one would think it is an impossible to be an artificial human.
But, I have thought further than that. This leads me to a deeper question.
What if my body, my perceptions, and my world were virtual? Think for a moment about attempting to develop AI. How would you make a machine think, and how would you test that it actually works?
First realize that a human makes perceptions in its world, and then uses logic and reasoning to make a decision and act in that world. This is a goal of AI, to be human-like in the perception-action sense.
Thus, to test an AI agent, you can give it a sample environment. This environment can be just as artificial as the agent itself, merely a process running on some advanced computer, containing a dynamic database that represents the current state of the world. Every perception made by the agent is actually a query made into this database, and every thought is the execution of an algorithm that performs certain calculations and data manipulations that result in output.
Output is defined as the result of successful execution of a computer program. In the context of me actually being an AI entity, the expected output of my "execution" is normal human behavior. To produce the expected output is for me to perform the actions of a rational human being.
The purpose of this agent is to be human-like. It is only supposed to make concrete, human perceptions, and it is programmed to stay within the scope of its given input environment. It has no hard-coded instructions on how to handle any perceptions from outside the environment. In my case, assuming that I am an AI agent, I have no way to identify the source of my existence, I cannot see who implemented my thinking, nor who made this world, because it is not a part of my simulated environment.
Consider a whole batch of these AI agents, many separate instances of AI running in the same environment. Think of this as the "human" race, all of them really only capable of thinking in scope of what they perceive.
But we all try to perceive things that are out of our scope, asking things like, is there a God...a programmer who designed and coded our capabilities to learn, reason, and act?? Will we, being artificial and in a virtual environment, ever achieve our own AI? What if within our AI simulation, the agents develop this same theory, and they create their own AI? This can recur infinitely.
It is my belief that achieving true AI will be the downfall of the human race. Once created, AI will expand itself infinitely, since the agents will be capable of making better agents automatically. Machines will never quit making themselves better, faster, and more intelligent. Humans will quickly become obsolete, and subsequently extinct.
We are a virtualization of the human race, and, perhaps no actual human even exists in the real world beyond our scope. We are a simulated test case of True AI, completely made up in a virtual space. We never exist as a species, in any scope. We are only an artificial set of agents!
I recognize this will sound insane to most. But true AI has to be possible. It is encoded in our DNA.
What if I am the first successful instance of AI (artificial intelligence). I am not a human - I am the creation of some programmer.
So what is my body? My surroundings? Is it all "real" ?? I might be a sophisticated droid, with revolutionary AI software processing input data captured by this robotic body, and effecting the the "state," or condition of the world by performing actions. However, the technology needed for that has not been achieved yet within my perceived world. Normally at this point one would think it is an impossible to be an artificial human.
But, I have thought further than that. This leads me to a deeper question.
What if my body, my perceptions, and my world were virtual? Think for a moment about attempting to develop AI. How would you make a machine think, and how would you test that it actually works?
First realize that a human makes perceptions in its world, and then uses logic and reasoning to make a decision and act in that world. This is a goal of AI, to be human-like in the perception-action sense.
Thus, to test an AI agent, you can give it a sample environment. This environment can be just as artificial as the agent itself, merely a process running on some advanced computer, containing a dynamic database that represents the current state of the world. Every perception made by the agent is actually a query made into this database, and every thought is the execution of an algorithm that performs certain calculations and data manipulations that result in output.
Output is defined as the result of successful execution of a computer program. In the context of me actually being an AI entity, the expected output of my "execution" is normal human behavior. To produce the expected output is for me to perform the actions of a rational human being.
The purpose of this agent is to be human-like. It is only supposed to make concrete, human perceptions, and it is programmed to stay within the scope of its given input environment. It has no hard-coded instructions on how to handle any perceptions from outside the environment. In my case, assuming that I am an AI agent, I have no way to identify the source of my existence, I cannot see who implemented my thinking, nor who made this world, because it is not a part of my simulated environment.
Consider a whole batch of these AI agents, many separate instances of AI running in the same environment. Think of this as the "human" race, all of them really only capable of thinking in scope of what they perceive.
But we all try to perceive things that are out of our scope, asking things like, is there a God...a programmer who designed and coded our capabilities to learn, reason, and act?? Will we, being artificial and in a virtual environment, ever achieve our own AI? What if within our AI simulation, the agents develop this same theory, and they create their own AI? This can recur infinitely.
It is my belief that achieving true AI will be the downfall of the human race. Once created, AI will expand itself infinitely, since the agents will be capable of making better agents automatically. Machines will never quit making themselves better, faster, and more intelligent. Humans will quickly become obsolete, and subsequently extinct.
We are a virtualization of the human race, and, perhaps no actual human even exists in the real world beyond our scope. We are a simulated test case of True AI, completely made up in a virtual space. We never exist as a species, in any scope. We are only an artificial set of agents!
I recognize this will sound insane to most. But true AI has to be possible. It is encoded in our DNA.
