Did you know that there are different types of Artificial Intelligence?

The Artificial Intelligence it is already a reality. Every day there are new tools that use AI, but whatknow the different types of AI? In this blog we tell you everything you need to know of them and what purpose they serve.

First of all, the different types of Artificial Intelligence bring with them different ways in which we can try to make machines think and act as humans. It is an idea that always seemed futuristic and completely out of science fiction, but today we already have machines super able to perform tasks in our routines... although I do not know.

The appearance of the IA has helped us to perform tasks and execute sophisticated process of a much more simple. If we talk about digital marketing, there are tools such as the Spicytool, for example, that facilitate the access to the digital advertising without the need to be an expert in digital marketing. This is completely innovative.

In addition, in these times so dizzying, where many times we are “running”, the ability to automate various processes, is one of the main benefits brought to us by the Artificial Intelligence. It is a wonderful thing, because it helps the user to be more efficient and get incredible results and outstanding, which otherwise would have been much more difficult and hence impossible.

Types of Artificial Intelligence and how to apply

Artificial Narrow Intelligence (ANI)

So simple and making a parallelism, we can compare it with a person trained in a complex function, and will only be dedicated to that. It is a type of artificial intelligence more inflexible, that is to say that it is not attached to or adapted to different systems or machines, but instead focuses on a single work, and complex. 

The model ANI is prepared to act in one role, when projected its operation through the programming of their actions, are looking to reduce your performance as much as possible.

What is sought is that you can only play its role or task.

It's an Artificial Intelligence with character reagent and limited memory, different to the other types of AI that are designed to meet various and/or multiple functions.

Then, the ANI is not an artificial intelligence that can reproduce the human behavior, but can be made to simulate it. Remember, you are only geared to specific objectives.

Where is it used?

• Assistant virtual, such as Siri, Alexa, Cortana
• Facial Recognition
• To filter spam e-mails
• In systems of autonomous vehicles.

Artificial intelligence-General (AGI)

It is considered one of the most versatile models in that there is currently, as it is able to mimic human intelligence and has a high capacity for action. That is to say, it is capable of learning and based on that replicate actions to solve various problems or issues.

As it is an Artificial Intelligence that “thinks”, it's like they have a unique understanding to a situation, it does not act in a manner completely robotic. Face different problems, different solutions. Their ability to adapt to different scenarios is what generates that has an activity of resolution closest to the human mind.

In its theoretical basis, the AGI, has the ability to assess and detect those different needs, processes, and even emotions, to act the right way.

It is a unique feature in relation to the other AI.

Artificial superintelligence (ASI)

As the name says... The most powerful, because it is able to become aware and autonomous.

The ASI is not simply to replicate human behavior, but that you can think better and to be more skilled than us. It is basically the model that many times it is portrayed in the movies where robots dominate the world..

Obviously that's not going to happen, but if the idea that they may be super smart, just like movies.

It is an artificial intelligence that is still in development and improvement. And you need to take care in the development of the same, as it requires you to do a careful analysis of the possible consequences that might have. That's why you always work with responsibility.

Machine test

This type of artificial intelligence... machines Are the simplest and most ancient known. Obviously has a capacity that is much more limited than the models that we have been telling you. Are simple: they reproduce the human behavior in front of a stimulation. Action – reaction.

Do not have the ability to learn, only respond. Eye can be very useful, although they are slowly becoming obsolete. 

In the late ‘90s were tested, and left us a great moment when the machine created by IBM and call Deep Blueovercame the great Garry Karsparv in chess.

Limited Memory

As the model described above, are machines-reactive, but that do have a little bit of memory, which can learn from data.

Create small databases from their interactions, and can take small decisions in the face of a stimulus. 

Theory of Mind

It is one of the models of Artificial Intelligence which has been the subject of much work in recent years. Although it is not yet possible to know a lot about where it will be applied, or how far its development, it is said to be one of the most innovative. 

Looking for that can better understand the interactions, more precisely, the emotions, the need and reflection processes of the human mind.

This model, if it looks like something that is in the future, although there are many advances, experts say that there is still a lot of development in other areas of the studies of artificial intelligence.

Self-awareness

Just an idea that guides the development of artificial intelligence. There's still nothing concrete, but you have the idea that in the future the machines believe consciousness of themselves. Would be the highest level that you can achieve artificial intelligence.

That is to say, one has the idea that the AI can understand all of the emotions, to have your own and be able to understand in-depth what happens with the people that interact with it. It looks like something extremely ambitious, but it is the horizon that have the scientists, although it is difficult to know how many years it will take to achieve this.

In addition, we can not deny that it can be dangerous, since the machines would have a reasoning that can't be stopped all the time. Much work remains to be done.

Today's technology allows many machines to act as human beings, facilitating and being useful to provide solutions. It is a great time to incorporate in our efforts to achieve greater productivity.