What is Artificial Intelligence?

Artificial intelligence or AI is the science of making intelligent machines, as well as the engineering skill required to build the machines, according to John McCarthy of Stanford University’s Computer Science Department. Intelligence can be found in many different animals as well as human beings. The term was first coined by a team of scientists, including McCarthy, from Dartmouth, Harvard, IBM and Bell in 1955. Governments and regulatory agencies have not yet determined a complete definition of AI, but some forms of computer-based programs can perform problem-solving functions and can learn. Both processes are considered to be part of the definition of “intelligence.”

Artificial General Intelligence (AGI) vs. Artificial Narrow Intelligence (ANI)

Because computer programs and machines operate so differently from human beings or animals, research into what constitutes intelligence has been challenging. In the mid-2000s, scientists began to work on human-independent definitions of artificial intelligence. They focused on measuring a machine or computer’s ability to perform tasks in a specific, narrow range of functioning, rather than trying to force machines to fit into a more general ability to learn across the entire spectrum of thought and learning. Komodo, a chess program, can play chess at a much higher level than any human. The program is considered highly intelligent at chess, but it can’t transfer that ability to other areas. Komodo is an example of ANI, or artificial narrow intelligence.

IBM’s Watson

IBM’s Watson became famous for defeating human champions on the television game show Jeopardy. Watson is a question-answering computer program that uses natural language processing, as well as several other forms of information processing. The program incorporates machine learning to improve its ability to answer questions. Watson uses massively parallel processors and employs ninety IBM Power 750 servers, with 16 terabytes of RAM. It can search through the information contained in a million books per second. Following Jeopardy, Watson has been put to work helping physicians diagnose cancer patients, and now has a dedicated group at IBM, where it is partnering with numerous businesses and industries. New versions of Watson are in development to serve the pharmaceutical, biotechnology, and publishing industries.

Different Types of AI

Any non-human or non-living system that can function in an intelligent way could be considered a form of AI. A useful way to think about AI is a system that can perceive its environment, and take actions to maximize its chances of survival. AI involves research in computer science, philosophy, linguistics, cognitive science, probability theory, logic, and many other related fields. In addition to computational intelligence, learning abilities, and pattern recognition, scientists are developing artificial systems that can use “fuzzy” logic and biologically inspired programs that can grow, develop and change.

Artificial intelligence is controversial, in the sense that machines that can learn independently of human control inspire fear in many people. Some scientists and technology experts like Stephen Hawking and Elon Musk have warned that uncontrolled AI could represent a threat to humanity. Although IBM’s Watson can beat human Jeopardy contestants easily and is creating recipes for food enthusiasts and programs to help physicians diagnose rare types of cancer, the machine has not shown any interest in conquering the world, starting wars, or taking the average person’s job.

See also: 5 Online Resources For Learning About Competitive Intelligence