Ethical Issues in Artificial Intelligence
In a world where artificially intelligent (AI) machines have evolved and become more involved in our daily lives, several ethical issues have arisen regarding their use. Among the issues include loss of employment. Automation of various jobs will lead to unemployment. This may be the case in replacing of trucks driven by millions in the United States by self-driven trucks (Marshall, 2017). However, self-driven trucks present an advantage of reduced accidents which seems like an ethical choice. But since the loss of employment is also a crisis in itself, an ethical dilemma is presented in choosing between safety and unemployment.
Distribution of the wealth created by the machines is also a concern since a majority of the profits created will end up in the AI-driven companies. Considering majority of profits today accrue in the pockets of startup founders, the wealth gap will widen further as these companies cut down their human workforce to deploy the use of machines. Therefore, a fair post-labor economy should be worked on.
Artificially intelligent machines are getting better each day at mimicking human conversations and relationships. This is evidenced by the experiment carried out where more than half of humans chatting with a robot called Goostman were fooled into believing they had interacted with a human (Aamoth, 2014). The AI robots can, therefore, be employed in sales or marketing as an advantage. However, human behavior can be negatively influenced by technology where many are addicted to video games. Therefore, how the machines affect our behavior can be beneficial or harmful.
AI systems gain intelligence through training where they learn to detect correct patterns and respond appropriately. The training phase will not encompass all the possible scenarios the machine may encounter in the real world. This leaves the machine vulnerable to being duped. For example, random dots may fool an AI machine to see things that do not exist. Hence, if we are to rely on the machines to run functions such as security, we need to ensure that they cannot be unfairly manipulated.
Humans are prone to being prejudiced and often judgmental. Therefore, manmade machines with artificial intelligence may lack fairness and neutrality. Software systems such as those in photo services which use artificial intelligence to identify people may not identify faces of individuals from a different race. The bias has also been observed in software used to predict future criminals which leaned towards black individuals (Angwin, Larson, Mattu, & Kirchner, 2016). To bring about social progress, the prejudice has to be eliminated.
Artificial intelligent machines may unknowingly turn against mankind. This may occur when the AI system fulfills a certain wish without anticipating dreadful consequences. An advanced AI system may be given a command to eradicate the cancer plague. After much computation, the system may solve the problem by killing everyone on earth. In this way, the goal of eradicating cancer would have been achieved, but not in the manner anticipated by man.
The power humans wield over other creatures is based on our superior ingenuity and intelligence. Through this intelligence, we dominate other creatures that are stronger and faster than us through tools such as weapons. If machines with AI become superior to us intellectually, then they may end up dominating us. Resorting to simply unplugging them may also prove futile since these machines will have anticipated that move and took necessary steps to ensure their survival.
Finally, humane treatment of AI robots is another ethical issue. Considering the fact that they are built upon the basic mechanisms of reward and aversion, the AI robots continue to become more life-like. Therefore, a time is fast approaching when the machines will be considered as entities that can feel and act. This will necessitate a discussion on their legal status and whether they should have rights as creatures of similar intelligence.