Can we verify if AI in our homes are smart?
“You may not realize it, but Artificial Intelligence is all around us.”
-Judy Woodruff.
Now, can we make connections? Is AI all around us? You may not know it, but it is. The real question that arises around the topic of AI is “Are all the AI around us really that smart?” An example of a type of AI around us today, maybe even in your home, is Artificial Narrow Intelligence or in other words, Weak AI.
What is ANI (Artificial Narrow Intelligence)? It is the type of AI you are currently using to read this article, your phone, tablet, or laptop. An example of this would be IBM’s Deep Blue, a reactive machine that beat the Chess Grandmaster, Garry Kasparov, in 1997. Another example, which is extremely relevant in today’s world is Google Home, many of us may have this in our homes and could be asking it questions at the very moment. You may think this machine is far developed in terms of AI, but not quite. If you ask your Google Home a question, all it is doing is responding to you with a pre-programmed answer. Therefore, as software updates approach, the variety of answers for each question increases. Giving the users more diversification in the answers provided to them.
You may have a question rising, how can this be proven? Simple, I did a test on my Google Home to see if it changed its responses when I asked three of the same question in different ways. I asked, “How are you doing?”, “How are you feeling?”, and “How is everything?” and the response every single time was, “I’m doing great thanks for asking, what can I help you with?” I am using this as an example because of how it differs from asking a human being the same questions in different ways. Usually, we will receive three different responses, one for each question that is different from the next. Because our brains are of higher intelligence than those of Weak AI machines, we tend to respond more emotionally and logically than machines, which is why our response is different every time. Even when people ask us the same question in different ways we understand the slight difference, and that allows our brains to develop different responses. This is something that Weak AI machines do not have the ability to do, yet. You can now prove that the AI in your home is probably not the smartest machine yet because of the pre-programmed responses.
Coming back to reactive machines, such as IBM’s Deep Blue, they are one of the most basic machines and one of the most prominent examples of AI. Another category in ANI is limited memory. Limited memory has the capability of learning from previous data, similar to when we humans learn from our mistakes. Thus, modern-day AI can be categorized under limited memory as it keeps learning and developing from previous data.
Does this indicate progression? Yes, every little step we take to make this world a better place is progression. From reactive machines to limited memory machines, this is tremendous progress. However, we have not achieved the most advanced AI yet. Currently the progress is roughly a third of the way through, there are still two types of AI to achieve, Artificial General Intelligence and Artificial Super Intelligence.
