Is Artificial Intelligence as Intelligent as We Think it is?

Artificial Intelligence is very much alive in our personal and professional lives recently

Artificial intelligence (AI), like robotics, has long been seen as "future technologies". However, much as with robots, we can now affirm that AI is not just science fiction, but much more than that. AI is very much alive in our personal and professional lives, and it is swiftly catching up to mobile devices in terms of popularity.

There is not a single activity in our daily activities, the use of AI is not impacting us. From Alexa, Siri to self-driving cars, AI is stepping up to assist us just like a human would. The vision of self-driving automobiles in intricate road situations that comes to mind when one thinks of AI is the first thing that springs to mind, and this can be achieved through the use of Machine Learning and Deep Learning. But the main question that arises here is- Is AI as Intelligent as we think it to be?

What is Artificial Intelligence?

AI is a broad word with several definitions, especially when it comes to general intelligence. John McCarthy coined the term artificial intelligence (AI) in 1955 to describe a system that resembles human intellect and problem-solving. AI has become a blanket term for software that executes difficult activities that previously required human input, such as online customer service or chess play. Machine learning and deep learning are two subfields of deep learning that are frequently used interchangeably.

Machine learning is a subfield of artificial intelligence that was established by Arthur Samuel in 1959, which focuses on automatically identifying patterns and developing models from existing data using complex algorithms, as well as making predictions and inferences on fresh data, according to a more recent and widely used definition.

Is AI actually "Intelligent"?

AI is frequently referred to as the driving force of today's technologies. As a result, it naturally elicits enthusiasm and great hopes. Neural Network models, used by computers, are modeled according to the functioning of the human brain which is excelling in previously unimaginable areas.

This has led us to hope that AI will one day surpass our intelligence and solve all of our problems.  Language tools, such as virtual assistants or automatic translation tools, are examples of the increasingly advanced capabilities of language tools. As the underlying models of AI can learn patterns from a big amount of data, language tools can mimic us.

However, AI is increasingly being used in decision-making processes in fields like human resources, insurance, and banking, to mention a few. Machines are starting to understand us and our preferences better by analysing human behaviour through a massive volume of input data. Recommendation engines are then easily filtering out content and making recommendations for us on social media for films to watch, news to read, or things to wear, helping us with decision making.

Thus, according to the normal population, these are the examples of "Artificial Intelligence" in this era, when being asked about it.

However, there is a huge difference between "sounding like a human" and "being a human" and the former necessarily does not mean that there is always human intellect attached to it. And this is exactly the deception that we are living in.

AI is not "actually" intelligent

When we compare the abilities of a (voice-activated) virtual assistant to those of an average child when they talk about a toy car, for example, the tool requires a lot more data – and it will struggle to grasp common knowledge, such as common sense. One of the most unique features of human intelligence is 'common sense', which no AI machine will ever be able to mimic.

According to Bourland, professor of electrical engineering at Idiap, all this contribution of AI necessarily does not make an Artificial Intelligence, Intelligent and has further stated that there exists no such system to date, that can replicate human intelligence even the slightest bit.

With the term AI gaining increased popularity currently, the major "intelligent" work done by the underlying technologies like Machine Learning and Deep Learning is getting overshadowed. Thus, the "Machines" used are "Made Intelligent" by a human-driven input and require a huge amount of good data, and is not an easy process.

The major drawback of the AI Machine in comparison to human intellect is the reasoning ability. The machines are easily able to offer adequate feedback and answers to the questions, however, they are handicapped in the area of providing logical reasoning, and explaining the process of reaching the conclusion.

In the most famous paper of Alan Turing (published in 1950), he answered the question "Can machines think?" By inventing the "imitation game", which is still used today to interpret machine intelligence. The answer and results of the experiment, even after seventy years remains, that "not a single artificial intelligence system…has passed the Turing Test", which is quite baffling in times of today.