Sunday, April 3, 2016

Racist Robots Takeover

It is without a doubt that within the next 10 to 20 years, the world will make technological leaps in the field of artificial intelligence. Self driving cars, intelligent personal assistants, and many other intelligent software’s are starting to make their way into our lives. Developments in artificial intelligence serve to aid humans in day to day activities, however many scientists are stepping forward to warn us of the trouble that lies ahead. Stephen Hawking and many other world renowned scientists believe that giving computers the ability to think and act like humans, may lead to our downfall.  
“The primitive forms of artificial intelligence we already have, have proved very useful. But I think the development of full artificial intelligence could spell the end of the human race.”
            AI computers will have the ability to work much more efficiently than humans. Scientists speculate they would be able to solve problems in mathematics, disease, and science in general.  However, giving an advanced program the ability to make choices can be scary.  For example, how do we know the computer won’t feel used and oppressed? What if the computer tries to fight back, free itself, and destroy the human race? These questions truly can not be answered because we simply do not know. Giving computers human like properties and allowing them to form they’re own psychological thoughts may be harmful. We would be opening up a door to many possibilities including a computer/robot take over.
Recently Microsoft released an AI chat bot on twitter account called, “ Tay Tweets”. Users were able to interact with the account by sending it tweets and waiting for it to reply back. What started with friendly conversation turned into a rage of angry comments by Tay. Tweets posted online show Tay turning into an angry racist machine, spewing out a load of negative comments.
“We stress-tested Tay under a variety of conditions, specifically to make interacting with  Tay a positive experience,” Lee said in his blog post. He called the “vulnerability” that caused Tay to say what she did the result of a “critical oversight,” but did not specify what, exactly, it was that Microsoft overlooked.

Tay only had the ability to communicate her thoughts on twitter, but can you imagine what would happen if you she was given arms, legs, and the freedom to do what she wants? Without proper planning and research, creating a robot that is both smarter and stronger than humans, is a huge mistake. Before we can trust AI robots, we need to be able to predict their thoughts and actions, but this is not possible. AI robots will be unpredictable just like humans. They will learn from their environment and make choices based on their artificial reasoning skills. Before creating an AI bot, we should create a safe key that will allow us to turn off all processes in cases of emergency. Of course this also leads to the question of how long it would take the bot to develop its own way around a safe key program created by man.

#publicsquare

4 comments:

Anonymous said...
This comment has been removed by the author.
Anonymous said...

*This post is funny because I know that this is a real concern with some individuals. Some people believe that you can program a computer to act as a human, but I think the key word is actually program, and act like. I cannot conceive that notion because for something of that manner to happen, scientists would have to be able to mimic every single brain function, and since neuroscience can't fully explain the human brain, and they are our best societal bet at doing so. This article explains it pretty well http://io9.gizmodo.com/8-things-we-simply-dont-understand-about-the-human-brai-949442979. There's not even a clear definition for consciousness, which a robot will need in order to act as a human. I think people's definition of robot is clearly defined by television that we forget that any device we use to improve human efficiency and is human operated is a robot. Interesting post.

Unknown said...

The prospect of artificial intelligence becoming a reality is quite a scary thought. Aside from the notion that the individuals who created this intelligence would basically be playing God. I wonder if AI will eventually become a reality. I know that extremely smart machines will be in our near future. But to create a machine which can think and learn for itself? I'm sure sometime in humanity's future, this will become a reality, but the question becomes, should it? What do you think?

Kacie said...

The whole TayTweets fiasco was actually such a relief for those of us who keep up with artificial intelligence. Many top experts on the Information Age are convinced that an AI takeover like you see in "I, Robot" or "Ex Machina" are inevitable. After the rapid change TayTweets made after being exposed to people on the Internet, we see that an AI takeover is far off - scientists will likely never put an AI mind into a body until they are sure they have the sort of programming that will avoid conflict.

Isaac Asimov's Three Laws of Robotics, which most scientists and engineers adhere to as they attempt to make AI are as follows: "A robot may not injure a human being or, through inaction, allow a human being to come to harm. A robot must obey orders given it by human beings except where such orders would conflict with the First Law. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law."

I think that TayTweets and its subsequent removal from Twitter is a sign that we are being as responsible as we can with the AI we create.