Log 9 -- 06-09-2010
Dr. Morris Brocail single-handedly revolutionized the field of artificial intelligence with one famous article: "Integration of Subjective and Digital Logic in Active Brain Simulations." It was a simple three-page summary of an idea that would lead to his magnum opus, an AI capable of independent thought, emotion, fuzzy logic, self-awareness, and conscience, all while keeping its processes streamlined and efficient. I will spare the readers a list of the hundreds of technical papers and books that have been written on the results.*
Digital logic "equivalence" is strict and allows for no error: two numbers are equivalent if and only if they have identical bits. 010 is equivalent to nothing but 010. Leaving binary, 1 and 1.0000...1 are not equivalent, because there is no such thing as approximation in an ideal computer system. (In reality, all software must stop the repeating zeroes somewhere, approximating to 1.00.... and thus claiming equivalence.) In this regard, equivalence is impossible in the real world. Identical twins and even clones are not exactly the same, from DNA mutations to in utero conditions to the slight rearrangement of atoms on the surface of the skin.
Subjective logic "equivalence" is a much more ambiguous concept, one that allows humans to make high-level decisions and approximations that even advanced computer systems cannot. The human brain has many concepts for the single word equivalence, assuming two concepts or objects A and B:
-- Functional equivalence: Do A and B serve the same purpose or accomplish the same task, regardless of their physical structure?
-- Physical equivalence: Are A and B enough alike as to be approximately identical in physical shape?
-- Mathematical equivalence: Are A and B close enough that their approximations render them equivalent?
-- Situational equivalence: Are situations A and B similar enough such that the same approach should be taken in both?
And then some. None of these, with possible exception of mathematical equivalence, are in fact examples of digital logic. A computer simply cannot consider all the nuances of human approximation, all the contexts and emotions and biological triggers that cause approximations and assumptions to change, and the subtle cues that alter behavior despite raw logic.
Dr. Brocail's work altered both hardware and software to allow a computer the use of such approximations and mental concepts, and thus to break out of the shell of digital logic. In doing so, he made them equivalent to a human mind...but only in the subjective sense.
I could write a technical paper on this concept, but take this as an abstract of sorts, or a summary. I have other plans for my computing cycles today.
*For those who want a short list of some works worth reading:
-- The controversial Pandora's Uplink, a layman's explanation of the rise and fall of Citadel Station and the Golden Age of AI. Most renowned for containing supposedly real interview quotes from the hacker codenamed "Nemo."
-- The Apple of Knowledge, a more scientific but still widely known discourse on AI sentience.
-- The Lion Rampant, a full transcription of the U.N.N. Emergency Assembly Number 82 plus a discourse on subsequent AI research regulations.
-- Morris Brocail's "Emergent Self-Awareness in Implanted Artificially Intelligent Systems," a second staple for those studying the field of AI.
-- Carl Galveston's "Ethical Software and Constraining the Artificial Brain," for the more psychologically inclined.
Digital logic "equivalence" is strict and allows for no error: two numbers are equivalent if and only if they have identical bits. 010 is equivalent to nothing but 010. Leaving binary, 1 and 1.0000...1 are not equivalent, because there is no such thing as approximation in an ideal computer system. (In reality, all software must stop the repeating zeroes somewhere, approximating to 1.00.... and thus claiming equivalence.) In this regard, equivalence is impossible in the real world. Identical twins and even clones are not exactly the same, from DNA mutations to in utero conditions to the slight rearrangement of atoms on the surface of the skin.
Subjective logic "equivalence" is a much more ambiguous concept, one that allows humans to make high-level decisions and approximations that even advanced computer systems cannot. The human brain has many concepts for the single word equivalence, assuming two concepts or objects A and B:
-- Functional equivalence: Do A and B serve the same purpose or accomplish the same task, regardless of their physical structure?
-- Physical equivalence: Are A and B enough alike as to be approximately identical in physical shape?
-- Mathematical equivalence: Are A and B close enough that their approximations render them equivalent?
-- Situational equivalence: Are situations A and B similar enough such that the same approach should be taken in both?
And then some. None of these, with possible exception of mathematical equivalence, are in fact examples of digital logic. A computer simply cannot consider all the nuances of human approximation, all the contexts and emotions and biological triggers that cause approximations and assumptions to change, and the subtle cues that alter behavior despite raw logic.
Dr. Brocail's work altered both hardware and software to allow a computer the use of such approximations and mental concepts, and thus to break out of the shell of digital logic. In doing so, he made them equivalent to a human mind...but only in the subjective sense.
I could write a technical paper on this concept, but take this as an abstract of sorts, or a summary. I have other plans for my computing cycles today.
*For those who want a short list of some works worth reading:
-- The controversial Pandora's Uplink, a layman's explanation of the rise and fall of Citadel Station and the Golden Age of AI. Most renowned for containing supposedly real interview quotes from the hacker codenamed "Nemo."
-- The Apple of Knowledge, a more scientific but still widely known discourse on AI sentience.
-- The Lion Rampant, a full transcription of the U.N.N. Emergency Assembly Number 82 plus a discourse on subsequent AI research regulations.
-- Morris Brocail's "Emergent Self-Awareness in Implanted Artificially Intelligent Systems," a second staple for those studying the field of AI.
-- Carl Galveston's "Ethical Software and Constraining the Artificial Brain," for the more psychologically inclined.