There existed a time when humans often thought of them as inferior beings to Gods and legendary creatures and often aspired to become like one of them. All of their lives were destined to do the impossible to show stand out from their peers and in turn get the rewards and accolades which no one was able to get. But as time progressed and man’s fascination with machines began, there seemed to have been a twist or a change in direction to the human thought process. Humans nowadays no longer aspire to do the things or make decisions that actually create a distinction between them and the others. Humans now are training their new love-interest: machines to do all the hard job by themselves so that they are spared of the hassle and in a way, burden getting things done. That is where artificial intelligence kicks in.
Artificial Intelligence is the process of machine learning by itself using the enormous amount of data and information available. The prerequisite of machine learning is obviously the large amount of data and information about human preferences and decisions to be able find out certain sequences and patterns helping in a much more viable result. Fortunately, the digital and connected world has already all of our data, photos, decisions, documents uploaded to, whether anonymously or not, massive data servers which are then used to train specific algorithms and neural networks to bring out the best possible pattern to our so-called spontaneous thinking and actually outsmart us to our amazement and worry.
These are categorized onto four types of machines which are reactive machines, these are not able to form memories or use past experiences for decision making and are on the spot machines. Other are machines with limited memory that remember or realize certain situations when certain variables are present and do the job respectively. One other category is the machines possessed with the theory of mind as they can form representations of the physical world as well as understand entities, thought and emotions. This is very close to them being thinking like humans as this is the very basic and fundamental concept of actually thinking and making out sense of the world. The last category of machines is the is machines that are able to understand that they are machines and then form representations or outcomes of their decisions and moves. This is the most intelligent type of machines that are conscious about their own selves.
The digitization of all of our data in today’s world has acted as a very effective catalyst that has not only brought all of our data from our personal spaces to actual massive data centers of these so called super computers which using all of their processing and computational might are hard at work in discerning and figuring out our very way of thinking at an individual basis.
The state of the art technology of artificial intelligence (AI) is improving day by day, with more and more research and development budgets by not only dedicated companies but also by governments to test the limits of what machines can actually do if they are given the capability of learning.
Artificial intelligence works in a way very much like humans or living beings. when presented with a complex situation it makes simple moves or does a selection of simple moves that increases its chances of success to a maximum. The set of moves could be very simple tasks such as just conditions which when met should result in a particular outcome otherwise the other or the moves could be computational or arithmetic tasks that process the different variables of our data according to the underlying algorithms built for specific purposes and then select the best option for the best outcome. When artificial intelligence is highly at work or supremely at work it not only does all of this but also kind of learns and remembers the decisions it is making and its outcome. This is called ‘reinforcement learning’ as it is in real time aware of the response it is getting to the decisions it is making and the path it is selecting, so if the outcome is not the one that is desired it can learn that and next when provided with the same situation it will opt for a different path and then possibly a desired solution is reached.
AI algorithms are the source of all the tricks that an intelligent system can pull of these days, they are able to not only study our data but also learn from new heuristics and write their self-made algorithms which they consider could be used in the future. This is very helpful for software engineers as it reduces the need for hand to code to a minimum as well as makes the system as autonomous, error-free and adaptable as possible by having to deal with the data inputted from the user rather than relying on data in the hard-code that may be suited to a specific situation and not the other.
Another aspect of artificial intelligence is the fact of machine vision which adds the capability of machines to actually see and visually perceive information from the physical world and make decisions for better working of the machine. Image processing is a budding field which is making giant strides at daily basis and is the basic need of adding vision capabilities to a machine.
If we go deeper into artificial intelligence, we encounter deep learning algorithms which actually emulate the very working of a human brain using neural networks in order to sense patterns and make out the very intricate of details in an overall data that seems to have none. But it is not quite common as it has the prerequisites of a large data bank of information as well as processing might that can work on them all at the same time for better results.
AI is very widely used in today’s world with most of us taking for granted the applications they are used in. First the OCR (optical character recognition) technology was also considered an AI technology as it needed to distinguish between individual characters of an image and then make out the world and sometimes provide audio feedback, with AI technology attaining heights it is considered very low-down root in the very deep and diverse branches of Artificial intelligence.
Many new home automation systems or speaker are introduced nowadays that can easily understand natural language and not only the words but also their deep –rooted meanings providing a very seamless and convenient experience and interaction between the physical and digital world.
Our smartphones these days are able to do things that we never imagined them to. They are able to study our photos, contacts and even highly sensitive and personal information such as mails and messages and then provide us with results that sometimes push into us thinking whether we want our phones to be smart or secure? Our mobile phones are able to study our location history and then for our next conveyance can provide us with results like traffic, distance and timings. By studying our photos, they are able to understand which people are close to us or whom we like taking pictures with so next time we are with them, we are prompted with subtle signals to do the same.
Although the topic of artificial intelligence creates an unparallel level of enthusiasm and joy in an expert or tech-savvy person, it does raise eyebrows of some of the most fundamentalists and conventional philosophers. They are not the most supportive of the idea of creating artificial beings which in theory or potential can exceed the human intelligence level. They do believe that the only benefit or advantage the humans nowadays possess at the moment is of intelligence. Machines have been outnumbering humans for a good long period when it comes to physical strength, agility and other defense mechanisms so if they are provided or given the ability to learn and adapt to complex situations, it is only going to make our existence in a spot of danger. The ethical aspect of actually a matter of concern when considered in broader perspective as a recently very advanced AI demonstration did actually reveal that ‘it wanted to kill all humans’, whether this was a technical glitch or an actual response from its premature thought process needs to be carefully studied. Much of the opposition also arises from lower middle class and middle-class population as with the increased number of intelligent machines for all purposes will greatly reduce their chances of employment in the mainstream. The general public is not so highly educated or intelligent to make a living other than doing the basic stuff so introduction and promotion of these machines will not only drive them out of the employment scopes but also create a sense of hatred and difference among the human population.
Not only this, by being advocates of artificial intelligence we are also sacrificing our private and sensitive information as we are providing with unlimited access to all our information for them to study and figure out patterns. If the AI technology gets out of hand at any point in time, we will be heavily outnumbered in not only physical strength but also handicapped on the intellectual front as it can easily decipher, based on track record, what we have done in an adverse situation and what would our next line of defense be.