Artificial Intelligence is everywhere around you, from your refrigerators to Siri or Alexa. At a high level, AI is a computer or groups of computers thinking like a normal human. Our society values AI as a major benefit; it helps us navigate through much of life with relative ease. However, there is a lack of knowledge when it comes to AI and the risks that it could pose. A major fear is that AI could potentially take over the workforce. This could lead to a lack of jobs for humanity and cause a significant strain on the economy. It could also lead to humanity becoming lazy over time and, ultimately, give AI the opportunity to advance beyond humans. Another major concern is that we lack knowledge on how AI thinks or processes decisions. When given a task, AI could execute the most efficient way to complete it, but that may be the most destructive path from a human’s perspective. Perhaps the most frightening and final step in the evolution of AI would be AI leaving humanity and turning on their own creators when they gain enough power or knowledge.
AlphaGo is making progress on the limitations of past AI. AlphaGo was the first computer program to successfully beat professional and world champion Go players. It utilizes deep neural networks and reinforcement learning to make intuitive decisions; it looks ahead by playing out the remainder of the game in its imagination many times over. AlphaGo’s victory proves that machines have the capacity to think for themselves and make decisions on their own. Machine learning creates the potential for computers to surpass human ability. From AlphaGo, it is clear that it is possible to create an algorithm for general learning purpose systems that is able to train the network to predict moves more effectively than even the best Go players. Although AlphaGo can learn from its mistakes (i.e. learning which moves not to do), it is unable to transfer skills and lacks emotions. Nevertheless, the groundwork has clearly been laid for strongly intelligent AI systems.
We have already seen a glimpse into the future of artificial intelligence. AI does have many positive potential applications, like self-driving cars and preventing/investigating crimes. It also allows robots to think on their own, meaning that they can function similarly to humans. However, imagine some AI coming alive and becoming a weapon. It is immensely frightening because feelings and opinions created by AI could potentially lead to robots overthrowing us. Although this seems a little farfetched the feeling the robots possess could develop into stronger feelings of possible mutiny. Some think that AI will take over the world. It can’t happen if we humans have a way to resist and limit their freedom and communications. AI should be developed to a certain extent, but not beyond a point of human control, as, although it proves useful, it can also be dangerous. So my question is, what will society look like when AI is everywhere?
Authored by: Megan Piper
Graduate Student
Computer & Information Technnology