Potentially, Artificial Intelligence will prove to be the final technological achievement of the human race. The future of technological progress may be changed forever if we succeed in developing a form of artificial intelligence that is so advanced that it can create Artificial Intelligence beings with even greater intellect. In terms of intelligence and accomplishment, such beings would be extraordinary.
A lot of people realize there are several kinds of AI, and some of them, like machine learning and natural language processing, are rather well-known.
In this article, we will discuss the basics of the four types of intelligence in AI: reactive, limited-memory, theory-of-mind, and self-aware.

Reactive Artificial Intelligence
Reactive AI, the simplest form of Artificial Intelligence, is pre-programmed to produce a predictable result in response to some input. In the same way that humans never learn from their mistakes, reactive machines never adapt to new conditions.
While the emergence of reactive AI marked a significant milestone, current generations of reactive AI are limited to performing only the tasks for which they were programmed. Thus, they are naturally constrained and open to development. From this premise, researchers created the next generation of artificial intelligence.
Limited memory Artificial Intelligence
Machine learning models that predict new insights from existing data or events are what make up this limited memory. When it comes to acquiring information, limited-memory machines observe the behaviors of their owners or the data they are given in order to learn from their mistakes.
Although temporary sample pieces of information are built on top of observational data and the pre-programmed data the machines currently hold, this advantage is short-lived. Autonomous vehicles are an example of a technology that has limited memory yet is already in use.
Autonomous vehicles, often known as self-driving automobiles, apply the limited memory concept by relying on a mixture of experiential and theoretical understanding. Self-driving cars observe their surroundings, look for patterns or changes in external factors, and make required adjustments so that they may safely drive and interact with other vehicles that rely on humans.
Theory of mind
The ability of an artificial intelligence computer to attribute mental states to other entities is referred to as “theory of mind.” The psychological origin of the phrase highlights the need for AI to infer the aims and objectives of entities (e.g., their beliefs, emotions, goals).
Using speech, image, and other data, emotion AI is being developed to recognize, simulate, monitor, and appropriately respond to human emotion. However, this is still a long way from being a theory-of-mind Artificial Intelligence, which can not only adapt its treatment of humans based on their emotional state detection but also understands them. This would be very useful in advertising, customer service, healthcare, and many other fields.
Yes, understanding as it is commonly understood presents a significant challenge for AI. An artificial intelligence that can create a great portrait has no idea what it has painted. The system can produce lengthy writings without comprehending their content. If Artificial Intelligence ever gets to the point of having a theory of mind, it will no longer face this obstacle.
Self-aware Artificial Intelligence
The last type of artificial intelligence is the self-aware. When this happens, machines will not only be able to read the emotions and mental states of other people, but of themselves as well. When AI achieves self-awareness, it will be on par with humans in terms of intelligence and consciousness, and it will feel and respond to emotions and desires in the same way that humans do.
Because we lack the necessary technology and algorithms, this type of Artificial Intelligence has not yet been built successfully. When we get there, some wonder if machines will be able to take over humanity because of this artificial superintelligence.
On the other hand, they might end up assisting humans in the creative process and working together.

Conclusion
For the time being, it is essential to concentrate on mastering all the four types of intelligence in AI. Moving through the many levels of artificial intelligence too quickly and carelessly could possibly be harmful to the development of Artificial Intelligence in the years and decades to come.