-
3 BCE
Talos, The First Concept of A.I
The image of an artificial being that performs human tasks first appeared in Ancient Greece. In Greek Mythology, Talos is a giant automaton created by Hephaestus, the Greek god of invention and blacksmithing. Talos was sent to protect the island of Crete from invaders. At his core, the giant had a tube running from his head to one of his feet that carried a mysterious life source of the gods called ichor. The story appears in the epic of Jason and the Argonauts, dated from the 3rd century BC. -
1943 - Creation of the concept "Neural Network"
Human consciousness is one of the main philosophical problems in history, with scientists seeking to understand how rational thinking operates and whether it can be replicated mechanically. The first research in this direction could only be done with the emergence of computers. In 1943, Walter Pitts and Warren McCulloch described the functioning of artificial neurons capable of performing logical functions, a concept that would come to be called "neural network". -
1950 - The "Mimic Game"
In an article published in 1950, the father of computing, Alan Turing, speculated that one day machines would have the same processing power as the human brain. That's why he proposed the Turing Test, or "Mimic Game," a way to test the intelligence of a machine compared to human cognition. -
1956 - Beginning of AI
The term "artificial intelligence" was coined in 1956 during a summer conference at Dartmouth in the United States. The event was promoted by computer scientists Marvin Minsky and John McCarthy, and brought together names such as Claude Shannon (creator of information theory) and Nathaniel Rochester of IBM to discuss ways to artificially replicate the human ability to solve problems. -
1974 - AI "Winter"
A Conferência de Darthmouth tornou a IA um campo da ciência e influenciou investimentos na área, inclusive por programas militares, como o Defense Advanced Research Projects Agency (DARPA). Mas o otimismo deu lugar à descrença, principalmente por causa da limitada capacidade de processamento e dos altos custos dos computadores. Fundos de financiamento para a área começaram a esgotar, e o ano de 1974 ficou marcado como o início do Inverno da IA. -
1980-1987 From the New Boom to the Second AI's Winter
A new boom in artificial intelligence began in 1980, with the popularization of Expert Systems, software with the ability to simulate the reasoning of experts in some field of knowledge. This type of AI began to be adopted by private companies and funded by governments, such as Japan's. This ended up creating a bubble, which burst in 1987 and started a second "winter". -
1996 - AI First WIN
One of the modern milestones of AI was the defeat of chess champion Garry Kasparov by the IBM computer Deep Blue. The feat represented the definitive proof of Moore's Law, proposed in 1965, which predicted exponential growth in computing power every year. -
XXI CENTURY - GPT-3.5
Starting in the 21st century, new concepts such as big data and machine learning, coupled with an increase in data processing capacity, have made AI flourish more and more. Today, we can already glimpse the abilities of general artificial intelligence to solve any type of task that a human can also accomplish. The launch of tools such as GPT3.5 in 2022 by DeepAI is an example of how this technology has evolved since the Old Era and how there is still much to learn and accomplish.