Discover the four waves of Artificial Intelligence
Posted: Sun Jan 19, 2025 10:21 am
If you talk about Artificial Intelligence with your network, you've probably already experienced this situation: for one person, such technology is something that belongs to the distant future - while for the other, AI is already a thing of the past. Confusing, isn't it? The truth is that, although it may not seem like it, these two people are talking about different subjects.
There are four waves of Artificial Intelligence - some that have already arrived and others that are yet to come. Understanding the different characteristics of each of them is essential to understanding their impacts on the economy and society. Mainly because, according to data from McKinsey & Company, AI will generate 13 trillion dollars for the global economy by 2030.
In an exclusive class for LIT students, Adriano Mussa, our ghana whatsapp data Academic and Artificial Intelligence Director and author of the book Artificial Intelligence - Myths and Truths; The Real Opportunities for Creating Value in Business and the Impacts on the Future of Work , explained how each of the AI waves can be defined.
Discover the characteristics that define the 4 waves of Artificial Intelligence:
1st wave: Internet AI
Believe it or not, the first wave of AI began in 1998 in the United States and spread throughout the world in the following years. It is characterized by the application of deep learning to data from internet users. In other words, the system was able to interpret and generate insights from a large amount of information. This is how companies like Amazon and Google work, since they have reached a level where they understand each customer so well that they know what and when to recommend a product, generating more sales.
This is the Artificial Intelligence considered "old" by some people because we have been living with it for some time. "Today this model is very mature in the US and China. In Brazil there are still many opportunities to use AI from the internet", added the professor.
2nd wave: Business AI
Business AI began around 2004. In this case, deep learning is also used, but this time it is applied to a company's data. In addition to being able to obtain insights that are difficult to see with the naked eye, the machine also learns to perform tasks such as credit assessment, fraud identification and even medical exam diagnoses. Therefore, when we think of Business AI, we are often talking about the unprecedented democratization of certain services.
To use this type of technology in business, it is not necessary to have a large company, just a good database. "Totally new opportunities arise in sectors that have already been digitalized," highlights Mussa.
3rd wave: Perceptive AI
This form of AI is less mature, but it is already being used in some places. So far, the technology is only able to read data generated by people while they are using a computer or other device. But what if AI could read and interpret actions and emotions offline? The idea of Perceptive AI is to give the machine eyes and ears. To do this, sensors are used using the Internet of Things , that is, objects that connect to a network and generate data.
An example of this is the Amazon Go store, where you don't need to go to the cashier to pay. The cameras can read when you pick up a product or not, and credit it to your account. In addition, it also generates data about your reactions to a product so that it can improve it and, therefore, generate more sales.
You may be wondering: wouldn't it be an invasion of privacy to generate data about people when they are offline? Professor Mussa says that there are several discussions on the subject, but it is necessary to put them on the agenda quickly, since technology advances without waiting.
4th wave: Autonomous AI
If in the third wave we gave the machine eyes and ears, in the fourth we give it arms and legs. "This wave is the last to arrive, but it is the one that will change our lives most profoundly," explained Mussa. The idea is to make robots autonomous, that is, capable of making good decisions.
Although the algorithm is already ready, the hardware that would give precision to the machine's movements is still being improved. Several tests are also needed for the system to learn how to behave. This is the case with autonomous cars, which are still undergoing training to generate the necessary data and learn how to drive while avoiding accidents.
There are four waves of Artificial Intelligence - some that have already arrived and others that are yet to come. Understanding the different characteristics of each of them is essential to understanding their impacts on the economy and society. Mainly because, according to data from McKinsey & Company, AI will generate 13 trillion dollars for the global economy by 2030.
In an exclusive class for LIT students, Adriano Mussa, our ghana whatsapp data Academic and Artificial Intelligence Director and author of the book Artificial Intelligence - Myths and Truths; The Real Opportunities for Creating Value in Business and the Impacts on the Future of Work , explained how each of the AI waves can be defined.
Discover the characteristics that define the 4 waves of Artificial Intelligence:
1st wave: Internet AI
Believe it or not, the first wave of AI began in 1998 in the United States and spread throughout the world in the following years. It is characterized by the application of deep learning to data from internet users. In other words, the system was able to interpret and generate insights from a large amount of information. This is how companies like Amazon and Google work, since they have reached a level where they understand each customer so well that they know what and when to recommend a product, generating more sales.
This is the Artificial Intelligence considered "old" by some people because we have been living with it for some time. "Today this model is very mature in the US and China. In Brazil there are still many opportunities to use AI from the internet", added the professor.
2nd wave: Business AI
Business AI began around 2004. In this case, deep learning is also used, but this time it is applied to a company's data. In addition to being able to obtain insights that are difficult to see with the naked eye, the machine also learns to perform tasks such as credit assessment, fraud identification and even medical exam diagnoses. Therefore, when we think of Business AI, we are often talking about the unprecedented democratization of certain services.
To use this type of technology in business, it is not necessary to have a large company, just a good database. "Totally new opportunities arise in sectors that have already been digitalized," highlights Mussa.
3rd wave: Perceptive AI
This form of AI is less mature, but it is already being used in some places. So far, the technology is only able to read data generated by people while they are using a computer or other device. But what if AI could read and interpret actions and emotions offline? The idea of Perceptive AI is to give the machine eyes and ears. To do this, sensors are used using the Internet of Things , that is, objects that connect to a network and generate data.
An example of this is the Amazon Go store, where you don't need to go to the cashier to pay. The cameras can read when you pick up a product or not, and credit it to your account. In addition, it also generates data about your reactions to a product so that it can improve it and, therefore, generate more sales.
You may be wondering: wouldn't it be an invasion of privacy to generate data about people when they are offline? Professor Mussa says that there are several discussions on the subject, but it is necessary to put them on the agenda quickly, since technology advances without waiting.
4th wave: Autonomous AI
If in the third wave we gave the machine eyes and ears, in the fourth we give it arms and legs. "This wave is the last to arrive, but it is the one that will change our lives most profoundly," explained Mussa. The idea is to make robots autonomous, that is, capable of making good decisions.
Although the algorithm is already ready, the hardware that would give precision to the machine's movements is still being improved. Several tests are also needed for the system to learn how to behave. This is the case with autonomous cars, which are still undergoing training to generate the necessary data and learn how to drive while avoiding accidents.