Many people have different ideas about artificial intelligence (AI). While some may think of AI as robots, others see artificial intelligence as a threat. AI can be difficult to understand since it’s complex and is constantly evolving.
In this article, we will explain what AI is and see how far it’s come by discussing the history of artificial intelligence.
What Is Artificial Intelligence?
Before we discuss the history of artificial intelligence, it’s important to briefly explain what artificial intelligence is designed to do. The main idea behind artificial intelligence is to allow machines to learn from experience. AI may be anything from a self-driving car to a computer that can play a game against a human component.
Much like humans, who become smarter the more information that they learn, AI is designed to “get smarter” by deep learning and natural language processing. Through processing information and recognizing patterns within the information, AI can learn specific tasks as well as (or in some cases better) the majority of humans.
Artificial intelligence isn’t simply just a smart computer or a machine that can accurately perform a task. Instead, it’s a broad field with a number of theories, methods, and uses various technologies, which may also include:
- Cognitive computing to help develop a more “lifelike” AI
- Computer vision assists in helping AI learn to recognize images and analyze the picture or video
- NLP (Natural Language Processing) allows AI, and human communicate using “everyday” language
What Artificial Intelligence Is Not
Although AI is designed to communicate more effectively with humans, think like humans, and even reduce preventable human errors (such as in AI vehicles), artificial intelligence is not a human and cannot replace humans.
While some sectors in the workplace, such as manufacturing, have use AI, some people see it as a threat to employment and even the overall livelihood of a community; artificial intelligence cannot replace some of the valuable skills needed in a variety of industries.
The History Of Artificial Intelligence
Now that you know a little bit more about what AI is (and what it isn’t), let’s take a look at the history of artificial intelligence and how it became what it is today.
When many people think of artificial intelligence, they think that the concepts stem as far back as the 1920’s or 1950’s, but some historians suggest that early AI concepts were thought about centuries ago.
Early Writings Discussing AI
Photo credit by: waitbutwhy.com
Although artificial intelligence didn’t become an official term until about 60 years ago, we can trace back to some of the first writings back to the early 1300s. It is also believed that the Ancient Greeks had myths about AI and the Chinese and Egyptians made mechanical devices that were designed to imitate humans (also known as automatons).
In 1308, Ramon Llull, a poet, and theologian, wrote The Ultimate General Art (Ars generalis ultima), in which he discussed his methodology for making paper-based mechanical means to create brand new knowledge. While Llull’s ideas are far from the AI we know today, his work is some of the earliest that was published using AI ideas.
Three centuries later, inspired by Llull and borrowing some of his ideas, Gottfried Leibniz, a mathematician and philosopher, wrote One the Combinatorial Art (Dissertatio de arte combinatoria). In his writings, Leibniz argued that human thoughts are nothing more than a combination of a small number of simple concepts.
The 1700s to Early 1900s
Photo credit by: mentalfloss.com
Ideas of artificial intelligence appeared Jonathan Swift’s popular work, Gulliver’s Travels, which was published in 1726. In his novel, describes an Engine on the island of Laputa. The Engine is said to be a parody of some of Llull’s early writings and ideas.
In 1763, Thomas Bayes created a framework for reasoning about the probability of events, which is referred to as Bayesian inference; Bayes’ ideas played an integral role in machine learning.
In 1898, Nikola Tesla, a well-known inventor, engineer, and futurist, demonstrated the world’s first radio-controlled vessel at an exhibition in Madison Square Garden. During his demonstration, Tesla said that his radio-controlled boat had a “borrowed mind.”
In 1914, the first chess-playing machine without any need for any human intervention was developed and demonstrated by Leonardo Torres y Quevedo, a Spanish engineer. While his invention is far from the computer-generated chess games we know today, it was a game-changing idea for AI.
1920s to 1950
Photo credit by: danielcschmid.com
Karel Čapek, a Czech writer, introduced the word “robot,” which derives from “robota” (which means to work) in his play, Rossum’s Universal Robots (R.U.R) in 1921. Four years later, the first radio-controlled driverless car was made by the company, Houdina Radio Control.
Six years after the word “robot” was developed, the first robot made an appearance in the iconic sci-fi film Metropolis. In the film, the robot destroys Berlin and terrorizes its residents in the year 2026. Metropolis is responsible for inspiring C-3PO’s look (from Star Wars), and the success of Metropolis most likely played a role of people becoming fearful of a future of robots.
Two years after the release of Metropolis, Makoto Nishimura designed Gakutensoku (which translates to “learning from the laws of nature”). As the first Japanese robot, it was able to move its hands, head, and make facial expressions through an air pressured mechanism.
In 1949, Edmund Berkeley published Giant Brains: Or Machines That Think, and his writings suggest that some machines are similar to a human brain, if a brain was made up of hardware and wire. His writings were in support of AI and suggested that machines had the ability to calculate and perform basic operations.
The 1950s and 1960s
Although the first chess-playing game, not requiring a human player, was developed in 1914, an article on developing a chess-playing computer program was published by Claude Shannon in 1950. The same year, Alan Turing wrote Computing Machinery and Intelligence which discusses the “imitation game” and is later known as the “Turing Test.”
A year later the first artificial neural network was developed using 3,000 vacuum tubes to simulate a network of about 40 neurons; the machine was named SNARC or Stochastic Neural Analog Reinforcement Calculator.
In 1952, the first computer program was developed to play a game of checkers, and it’s the first program to learn the game on its own.
In 1956, the term “Artificial Intelligence” was officially used at a conference at Dartmouth College. Over the next couple of years, engineers published a variety of articles on the programming language used in building and perfecting AI technology.
In 1961, Unimate, the first industrial robot joined an assembly line at a New Jersey GM plant and in 1965, Herbert Simon predicted that within 20 years machines would be capable of doing any work that humans can do.
Shakey, the first multi-purpose robot, was created to “reason” about its own action and in 1968, Hal, an AI computer, was a character in 2001: Space Odyssey.
1970s, 1980s and 1990s
Photo credit by: humaniod.waseda.ac.jp
During the 70s, 80s, and 90s there were many AI developments, fine tuning software, and perfecting ideas. Some of the highlights during this period include the first anthropomorphic robot, known as the WABOT-1, in Japan. WABOT-1 had limb control, vision, and conversation systems.
In 1979, The Stanford Cart, was able to cross a room filled with chairs, without any human help, in five hours. While this may not seem too impressive in today’s standards, it was the earliest example of how an autonomous vehicle works.
In 1980, WABOT-2 was built and was able to converse with humans as well as read and play music on an organ.
In 1995, shortly after the Web became easily accessed by many computer users in schools and other institutions, ALICE (Artificial Linguistic Internet Computer Entity), a chatbot, was developed by Richard Wallace.
In 1997, the first computer chess-playing program, Deep Blue, beat a world chess champion and a year later the first domestic pet robot, the Furby, hit toy stores everywhere.
2000 and Beyond
Photo credit by: accessholidays.ca
In 2000, Kismet, a robot that had the ability to recognize and even simulate human emotions was developed at MIT and the same year, ASIMO a Honda robot was designed to walk as fast as a human and deliver food trays to restaurant patrons.
In 2009, Google started developing their first self-driving car, which passed the first self-driving test in 2014.
In 2011, Watson, a computer who can answer natural language questions, defeated two champions on the popular game show (known for its smart contestants) Jeopardy!
As AI technology keeps evolving and is playing an integral role in hospitals, restaurants, factories, and even in hotels, the future is promising for both AI and humans to coexist and work together happily and more efficiently.