History of Artificial Intelligence: Key Milestones

Artificial intelligence (AI) is a set of sciences, theories, and methods whose main goal is to imitate human cognitive abilities. The history of artificial intelligence begins at the start of World War II, and its development is closely tied to the advancement of computer science, resulting in computers performing increasingly complex tasks that were initially only possible for humans.

In this article, we will explore the history of artificial intelligence so you can learn about its development and prospects, especially for those who want to study this fascinating discipline. Want to join us?

Early History of AI up to the 1960s: Emergence of the Discipline

The period between 1940 and 1960 was decisively shaped by the accelerated technological development driven by World War II and the desire to understand how machines could be used to ease human labor.

For Norbert Wiener, a pioneer of robotics, the goal was to unify mathematical theory, electronics, and automation into a single theory. Shortly before him, Warren McCulloch and Walter Pitts in 1943 developed the first mathematical and computer model of neurons. It served as one of the foundational references for subsequent developments in artificial intelligence, such as deep neural networks (which are currently on the rise).

Another early milestone in AI history is found in Turing's paper written in 1950: 'Computing Machinery and Intelligence'. In this paper, the British scientist described an imitation game in which a person had to distinguish whether their teletype dialogue was produced by a human or a machine. Furthermore, together with John von Neumann, Turing became a founder of the technology underlying the concept of artificial intelligence, creating a formal foundation for the architecture of modern computers and demonstrating that it is a universal machine capable of executing any program.

Alan Turing (1912–1954) committed suicide two years after being subjected to chemical castration for homosexuality. In 2016, the British government apologized and exonerated him.

Although Turing along with John von Neumann had a great influence on many subsequent developments, the conceptualization of artificial intelligence is mainly attributed to John McCarthy, who defines it as "the creation of computer programs designed for tasks that are currently performed more satisfactorily by human beings because they require high-level mental processes such as: perceptual learning, memory organization, and critical reasoning."

History of AI in the 1980s

Another key event in the history of artificial intelligence occurred in the 1970s with the advent of the first microprocessors. This door was first opened in 1965 with the Dendral system, and then in 1972 with the MYCIN system. Both systems were based on an inference engine programmed to become a logical mirror of human reasoning. After inputting data, the engine produced answers that demonstrated a high level of expertise.

This spurred the large-scale development of artificial intelligence, but it was only in the late 1980s that a small boom occurred in programming AI systems. However, maintenance and development were still too costly from an economic standpoint, and the results did not meet expectations.

ChipTest, the predecessor of Deep Blue, was developed in 1985 as part of a project at Carnegie Mellon University in Pittsburgh, Pennsylvania. Soon, in 1989, it was transformed into Deep Thought, a more powerful and efficient system.

Late 1990s: Deep Blue

In May 1997, the Deep Blue artificial intelligence system, developed by IBM, won a chess match against champion Garry Kasparov. Deep Blue was based on a brute force algorithm in which all possible moves were evaluated and considered.

The first matches between Kasparov and Deep Blue were played in 1996, with the winner being the one who won the most of 6 games. That time, the human won. A year later, after IBM improved its AI system, the computer was able to defeat the chess champion.

Kasparov's defeat is one of the key achievements in the history of artificial intelligence, its capabilities, and subsequent development from the first decade of the 21st century. However, Deep Blue actually only captured very specific features (the rules of chess), so it was far from modeling the complexity of the real world.

Big Data: The Boom in AI History

Two factors explain the new boom that artificial intelligence has been experiencing, mainly starting from 2010:

• Access to large volumes of data: previously, image classification algorithms were able, for example, to recognize a cat only if manual tests were conducted first. Today, a simple Google search allows finding millions of such images.
• High performance of graphics card processors. This speeds up the computation of learning algorithms. Tasks that used to take weeks to complete have been reduced to days, hours, or minutes thanks to the computing power of modern systems.

How did these two wonders come about? A complete paradigm shift of systems. The approach became inductive: now it's not just about encoding rules, but allowing machines to identify them independently through correlation and classification based on vast amounts of data: Big Data.

Unexpectedly, almost overnight, most research groups adopted this technology, which today offers a huge number of advantages. This type of learning has also allowed significant progress in text recognition, although we are still far from creating systems capable of understanding texts.