Some say that artificial intelligence is a threat to the human race. This statement is headline-catching but does not capture the complete truth. The real threat comes from complexity: as the world gets more and more complex, people delegate computer programs to make decisions for them. When the system gets even more complex, they delegate computer programs to deal with other computer programs. In a very complex system, which we shall arrive at soon, a break down in a small system could have unexpected consequences (the Amazon butterfly effect). As the world gets more complex, it becomes more fragile.
"The development of full artificial intelligence could spell the end of the human race. It would take off on its own, and re-design itself at an ever-increasing rate," Professor Hawking said. "Humans, who are limited by slow biological evolution, couldn't compete, and would be superseded."
"I am particularly concerned by the potential military deployment of robotic weapons systems – systems that can take a decision to militarily engage without human intervention – precisely because current AI is not very good and can all too easily force situations to escalate with potentially terrifying consequences", Professor [Mark] Bishop [Goldsmiths, University of London] said.
What threatens the society is complexity. The world has become more and more complex. With complexity increasing all the time, we have no choice but to rely on computer programs to make decisions on our behalf.
As we rely more and more on complex systems (such as the logistics programs and search engines), we struggle to understand them. Naturally some will produce more complex systems to help us deal with them. So the trend is that we rely on more and more complex systems.
So the problem is not due to artificial intelligence programs making decisions on our behalf. The problem lies in our need to delegate our decisions to computer programs. We have to do so due to he increase in complexity in the world.
Some find it convenient to call programs that deal with complexity artificial intelligence. It is also legitimate to question whether computer programs should be allowed to make decision for us, which they do everyday -- from search engines to how Internet traffics are routed. In fact, artificial intelligence programs do not have to make decisions for human beings - they can be there to find patterns for their users. So the statement that "artificial intelligence is a threat to the human race" is confused in several counts.
The trend is that the world will get more complex, which means we have to rely more and more on computer programs. With increased complexity, we shall soon reach a dangerous state where we don't know what the system will bring. In a very complex system, which we shall arrive at soon, a break down in a small system could have unexpected consequences (we are not adding sabotage to the equation). This is indeed dangerous.
Perhaps we should try to simply the world whenever we can!
[End]
All Rights Reserved