Ways to Counter Problems Arising From AI

There are many situations in which an AI could cause problems for the human race. Here are some solutions which could alleviate the danger.



Ways to Stop Direct AI Attack
In order to stop a direct attack by an out of control artificial intelligence, we could place a physical interrupt switch into its power supply, an “off” switch. If we lose control of the computer, we could simply turn it off. However, an artificial intelligence could be considered a new life form in which case pulling that switch would be tantamount to murder. You are shutting down a life, starving it to death if it were a human.

Another way in which we could control an AI would be to limit its ability to interact with the physical world. We could place it in an isolated computer with simple input and output channels not connected to any other machines. That way it could be utilized for its mental capabilities without the risk of the classic “building a robot army”. This solution has two problems. It limits the usefulness of the AI as it couldn’t interact with exterior machine elements. We couldn’t allow it to access the internet, limiting its reserve of knowledge, nor permit it to, for instance, run a factory. Both those abilities could permit the AI to gain a physical representation beyond our control. Second, this restriction infringes the qualifications of life. These are qualifications that all living organisms share. One of these is the ability to reproduce. We would be restricting this new form of life from creating more of itself, reproducing, which may eliminate the ability for us to call it life.

'''Isaac Asimov proposed a solution to the problem of the free will of AI. He created the Three Laws of Robotics. These laws were:'''

1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.

2. A robot must obey the orders given to it by human beings, except where such orders would conflict with the First Law.

3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

He later added a zeroth law:

0. A robot may not harm humanity, or, by inaction, allow humanity to come to harm.

These laws were hardwired into the minds of the vast majority of AI brains created in his novels. They are designed to protect foremost humanity then themselves. However, as illustrated in this article, they were not perfect laws. Their creator permitted to much room for interpretation. The AI minds got to determine what the laws meant, often to disastrous results. This example proves that no solution can be guaranteed.

Ways to Social and Economic Problems
The only way we could prepare the population of the world for this kind of a revelation, that life isn’t what we used to think it was, is to fortify people psychologically and secure particular jobs so that they cannot by law be performed by AI to keep some employment for humans. This section has too many variables to create proper security. People’s actions are often irrational and can’t be predicted. It is impossible to accurately guess what the world population would do under these circumstances in order to create a plan of action.

Questions for the Reader
1. Please consider the following questions and feel free to submit your responses to the comment section below.

2. Can you think of another way we could prevent an artificial intelligence from physically taking over the world should it choose?

3. Can you think of a method we could use to prepare the world for the cultural and social shock produced by the revelation of the creation of an AI?

4. Can you think of a way to prevent the economic damage an AI would produce should it be released to the industrial and commercial market?