As all of the world is shaking with the recent improvements of the works on Artificial Intelligence Technologies, we can observe how much we made progress by using Chat-GPT or other public applications that makes our live easier these days. As the researches go by, most of us has the same questions in our mind: Can AI reach the human capabilites one day? I’m sure a lot of us have watched these science-ficture movies with the concept of ridiculously-improved intelligent robot rebels fighting the human race (Terminator, Matrix , etc.) Can AI really become a threat for us as it advances rapidly day by day? Can AI really think and seek for freedom and power like real human beings? Can AI really think like a human? Most important of all these questions, can AI really become like one of us, like a real human being?
As the AI improves, as the lines of code and inputs becomes larger in size and number, its capabilities wides too. It gains capabilites of gaining more knowledge about more complicated things, like human interactions and emotions in general. As it’s ability to learn and adapt develops, it becomes able to impersonate a regular person and act like one. It can even even act like it has real emotions and understand what u feel with the way you talk. For example, a website called CharacterAI – developed by two ex-Google employees- gives you access of hundreds of specalized text-based AI that states on specific fictional or real life characters that you can have any type of conversation or do anything that you want, for example, you can play chess with Donald Trump. This shows us that AI can already used to how to communicate with human in many ways very close to a person to person communication. So, why can’t it become a real human in future? It looks like there is nothing prevents AI to become a real human, right?
Actually, there is. Although AI can learn and impersonate humans very well, but -as it is said- it can only “impersonate” what humans do and don’t and can’t really become a real one. When it encounters a situation that it can’t identify, it would be basically a fatal error for it. That, makes a difference between humans and AI. As you see from the example, AI is incapable of interpret new things and can’t react to it unlikely humans does thanks to instincts. Instincts are something that cannot be learnt, so AI can’t have instincts. For example, AI can’t react like a human in unexpected situations as it can’t have any fight-or-flight instinct, or can’t really have any real have any real feelings and reactions like human as it can only learn how to “impersonate” on learned sitouations. So, AI can do well at acting as a real human itself, but cannot become a full one without and instincts and real feelings.
To summarize, AI is improved to the point that it can learn human interactions in situations and use them to impersonate and act as a human. As it seems like it is enough to become a real human, it is clear that AI can really impersonate the feelings that we code them to and can’t react new situations as it has to learn it first. To answer the questions that we asked, no, AI cannot become a rebellious threat unless we don’t gave them the knowledge of how to rebel against humanity, so it is not a real threat as this technology is in good hands. No matter how much technology advances and how much AI becomes improved, human-like AI cannot be more than imitation itself.