The Limits of Artificial Intelligence
Photo by Matan Segev from Pexels
If you are here, it means that you are familiar with term artificial intelligence. Either you have read about it in school or have seen it in sci-fi movies or somewhere else. Talking about the limitations of AI, let me ask you one simple question first, do you know the definition of AI? You might be thinking to answer me with a yes, yes I know what is artificial intelligence. But what if I tell you that AI is a buzzword and it is almost impossible to properly define. It is this way because the definition of artificial intelligence is moving. People don’t call the things AI that they used to call. For example, a problem that seemed too complex to be solved by human and was solved by AI algorithm is no longer a problem of AI. Playing chess, is one of the examples. It was considered the peek level of artificial intelligence back in previous century. Now it hardly fits the criteria for AI. It is presented to the world as a super power that when given to a computer, it magically starts living like other living things. It start thinking. Well, that's one side of the picture. AI has its own restrictions and limitations.
Artificial Intelligence today is solving problems ranging from playing games against human players to assisting medical expert in diagnostics. (visit: 10 Powerful Examples Of Artificial Intelligence In Use Today.) We human are using AI everywhere no matter if we are aware of it or not. The dominance of big fishes in American tech market- Facebook, Amazon, Netflix, Uber and Google versus the rise of Chinese tech giants: Baidu, Alibaba, Tencent, has created fierce quest for resources and specialists. Yet, what are all of them racing towards? AI. Artificial Intelligence is the one thing, after military and financial power, which will decide which country the next super power will be.
However, today's artificial intelligence has certain limitations. It is not even near to the pinnacle of automation where human are. AI that exists today, is too far from achieving human level intelligence. Artificial Intelligence is compared with human intelligence because human is the most intelligent creature on earth and the goal of AI is to at least reach to that level of intelligence if it is not to cross. Whether it is a computer program describing a scene in a picture or an autonomous car driving automatically on busy roads and avoiding collision and pedestrians, AI always try to mimic human intelligence.
Although, it is admissible that today's AI performs better and often outperforms human intelligence for specific problem. For example, in financing, AI comes with better forecasting and decision. AI is helping experts in almost all walks of life including medical, military, business and law. But as far as generalization is concerned, it will take it no less that 50 years to achieve human level intelligence.
By generalization I mean human mind is a single organ, it learns from experiencing different situations. But a single AI model or algorithm can learn from only one kind of data. For example, an algorithm that has learned to identify vehicles in an image will fail to identify animals. Although, it could be trained on mixed data including both animals and vehicles but it would never outperforms human mind in that area.
Another biggest limitation of AI is that it learns from data. There is no other way for AI power machine to gain knowledge, unlike human learning. AI performs better than human only in those cases where enormous amount of data is available. It fails where creativity, critical thinking and logical reasoning is involved. Any AI machine would always be a machine which will do the job it is told to do. It can’t figure out how to solve a problem it has never seen before.
The reason for failure of AI at general problems is that human mind has a modular structure. Each section or module in human brain is responsible for different functions. For example, frontal lobe of human brain is responsible for speaking while occipital lobe is more concerned with vision. Deep learning and machine learning model lack this feature of modularity. One model is almost similar to a single module in human brain. Another reason for AI failing to reach human level intelligence is that human brain is the most complex structure in the universe. AI models have not gone that complex until today.