How to Understand What is & What is not Artificial Intelligence?How to understand what is and what is not Artificial Intelligence?


23 Views | Read Time 4 minutes |

“The machines surprise me very often” and that Alan Turing, considered the father of Artificial Intelligence (AI) and computing, has said it is an indication of the potential we can see in the short term. 

Birth of Artificial Intelligence

The concept of Artificial Intelligence was born in 1956, with a group of visionaries interested in building a machine, which was as intelligent as a human being and found successful. While presenting it for the first time in a session at the University of Dartmouth, as an example of a Super smart computer, it was capable of performing various functions.

Aroused such enthusiasm that science fiction writers, creatives and screenwriters from Hollywood exploited the possibilities, with visions from the most optimistic to the most worrying of what a world would be like where robots and humans coexist, however, in the decade of the seventies and eighties stormed the “Winter of Artificial Intelligence”.

Activities by Artificial Intelligence in the 80s and 90s

If it was one of the most tempting and promising ideas, nobody wanted to invest in research or technological development. The costs were very large, there were no tangible results on a return on investment. Although in the ’80s the Japanese government began to inject resources, it was until the arrival of the Internet and already in the mid-’90s when the term came back with more force. 

But what do we mean by “Artificial Intelligence”? To the possibility that the machines can perform complex actions, such as image recognition, process human language or make personalized recommendations for the consumption of products and services.

READ ALSO  European Football Clubs Are Turning to AI for an Assist

Although computers are still electronic devices that the only thing they can process are zeros and ones, their algorithms and all the processors of the last 10 years, such as GPUs (Graphic Process Unit), are what have made it possible for this category of intelligent algorithms can perform these tasks that until 10 years ago seemed impossible and were only conceived as part of science fiction.

Importance of AI

The most important feature of AI is that it is based on very complex statistical mathematical models, which allow it to make abstractions of large and complex volumes of data and that manages to divide them to a minimum. Each of these small data represents a microtask to statistically evaluate to generate a prediction. Conversely, the sum of all these predictions is what allows AI to perform these complex tasks. This is very important to understand and distinguish because there is a lot of misinformation about the ability to understand and reason, which until now are exclusive skills of human beings. The current AI, as surprising as the tasks it performs, is still an algorithm, that all it can do is process data and make a statistical decision. No computer has conscience or reasoning; They are simply capable of executing millions of tasks extremely quickly.

Classification of Artificial Intelligence

AI can be classified into two types: narrow or vertical, to designate those algorithms that allow solving a specific task with great efficiency. Here are autonomous cars, voice assistants, language translators, internet recommendation engines, etc. In this type of AI is the subcategory of “Machine Learning”, consisting of mathematical equations, which is described in the previous paragraph can identify patterns, and incorporate the results as part of the input data, to refine the model itself.

READ ALSO  See how an AI system classifies you based on your selfie

In other words, these algorithms evaluate the results and with them adjust the weights of certain variables to make more accurate future results, hence the “learn” analogy.

The other category of Artificial Intelligence is the so-called “General”, which does not yet exist, and refers to the possibility that machines develop something very similar to a human conscience that translates into autonomous decision making. 

Theory of Artificial Intelligence

In theory, the increase in hardware processing capacity and a combination of current AI algorithms could generate a reflective and reasoning capacity. However, the debate is very large and the main reason is that as we mentioned so far there is nothing that resembles this kind of intelligence.

The truth is the winter of Artificial Intelligence is behind us and we are sure that the surprise that Turing felt will continue to accompany us in this new technological stage. Without a doubt the limit is imagination. 



Source link

?
WP Twitter Auto Publish Powered By : XYZScripts.com