Artificial Intelligence: State Of The Art

Abstract

The birthdate of the field of artificial intelligence (AI) is open to debate. Its name, however, was bestowed upon it in 1956 by John McCarthy. From time to time other names have been suggested, such as “machine intelligence”, “cognology”, “theoretical psychology”, or “experimental philosophy”, but the original name endures. As the diversity of the proposed alternatives might imply, AI encompasses a broad range of activities and motivations.

Defining the scope of a scientific field is always difficult, usually resulting in something vague, broad, and uninformative; for example try looking up ‘physics’, ‘chemistry’, or’ biology’ in a dictionary such as Websters’. What is more, there are always areas, like molecular biology, that cannot be safely assigned to one field. For the front present discussion, I will offer the following description of AI:

Artificial intelligence is concerned with understanding the principles and building working models of intelligent behavior.

For some research workers, the first part of this description is paramount, and the goal is that of understanding the human intellect. They operate in the classical scientific manner, developing a theory and then building a computer model to test experientially, in much the same way as a physicist or a psychologist. For others, the second part is the driving force, and they seek to design machines that can relieve people of burdensome elect intellectual tasks. They do so by building computer systems that perform at an acceptable level of competence, and then they attempt to further understand and extend the methods and principles involved. This approach is more akin to aeronautical engineering and the development of aerodynamics. As in other fields, science and engineering, principles and applications, go hand-in-hand. These two main operational approaches share the very practical steps of the working model. For the time being at least, the model takes the form of a computer program, because the digital computer is currently the most powerful and versatile modeling medium available (as an evidenced by the decline of random logic in the rise of programmable logic in many areas of electronics ). Because of the centrality of the digital computer, AI is often considered to be a branch of computer sciences.

Whatever the motivation, the problems tackled by research workers and AI are typically large, complex, and involves ambiguity and uncertainty. AI attempts to come to grips with the real world, with all its confusion and richness, and to interact with people, with all their irrationalities and idiosyncrasies.