A “Man on the Street” interview asked people this question- what’s the most impressive invention of the last 150 years?” One gentleman suggested that it had to be the automobile. What else, he reasoned, had had such an impact, reshaping the very contours of the world? The second took a more modern approach, suggesting that the personal computer has had the greatest impact, putting immeasurable knowledge at our fingertips. The third man, however, was more down to earth. “What most impresses me,” he said, “is the thermos.” “The thermos?” the interviewer asked, “But why?” “Well, the gentleman explained, “some mornings my wife fills my thermos with cold iced tea, and when I drink it at lunch, it’s still cold. Other mornings she puts in hot coffee, and come lunchtime, it’s still hot. That impresses me because I’ve never been able to figure out, how does it know?”
Despite the wonders it is able to accomplish, the above funny highlights to real limits of AI. If it is artificial, it isn’t intelligence. If it is intelligence, it isn’t really artificial. Technology doesn’t truly know anything. Mercury may rise on a hot summer day or drop when the sun goes down but it neither knows what day it is, nor whether it is day or night. You can ask AI a question, but even if it gives you the right answer a million out of a million times, you can’t make it think. AI no more thinks than Rodin’s “The Thinker” thinks. Thinking requires consciousness, which is why Descartes made the philosophical Hall of Fame. “I think, therefore I am” isn’t the philosophers’ equivalent of “Live to Ride; Ride to Live” of Harley enthusiasts. Rather it is a profound affirmation that one cannot doubt one’s existence without first existing. In like manner, one cannot ponder the nature of one’s existence if one is created by mere man.
Dr. Frankenstein’s monster, Star Trek’s Mr. Data, Pinocchio may have seemed to puzzle over their own natures, but that’s because they are merely two dimensional literary creations. They didn’t ponder; their creators pondered for them. In short, they aren’t real. Neither is artificial intelligence real, if by real we mean something more than what men put in. In the same way that an animal is a higher order of being than a rock, so that which is conscious is a higher order of being than that which is unconscious. Humans will no more ever have AI overlords than God could ever have human overlords.
Such is not to say that AI might not come with some problems. (Though anyone with any economic sense should know that long term they won’t cost a single soul a single job.) Our concern, however, ought to be over actual conscious beings, those who are able to sin. Such might connect with AI, either those writing the programs, or those “speaking” through them. To put it another way, remember that we war with neither flesh and blood nor programs and algorithms but with principalities and powers.
I have read, I think in an article from the Discovery Institute, that they have put AI machines in an isolated room and let them chew on each other’s information & the results degrade into nonsense. In other words, to be relevant at all AI has to to have a source of constant creative thinking – human beings thinking otherwise things to not go well. Greatly enjoyed your post. You are correct, we need to be less worried about potential AI masters & more worried about the true holy God.
Thank you Mark. This piece was a bit of a challenge as, while I feel competent on the philosophical issues I don’t on the technological. Yet it seemed to me that the experts in the technological side weren’t so adept at the philosophical side.