Artificial Intelligence
Modern Artificial Intelligence (AI) is defined as the development of algorithms or computer systems capable of performing tasks, the same or better than humans. The ultimate goal is to develop systems capable of helping humans in getting their daily tasks done with ease. These tasks include generating or recognizing speech, making complex decisions, helping robots to work like humans, and even generating creative content. But as everything has some disadvantages, there are also pros and cons of AI. This article revolves around the topic of the “History of Artificial Intelligence”.
History Of AI
Early Stages (1900-1950)
In the early 1900s, scientists were inquisitive about the concept of the artificial human brain.
Many scientists wanted to know if it is really possible to create an Artificial Human brain. In this regard, many people found different terms for AI which laid as the foundation for Artificial Intelligence. Many scientists founded the basics of AI with different experimentations.
1912: Leonardo Torres Quevedo was an English chess player who built a chess-playing machine in 1912. For this, he was labeled as the 20th century’s first AI pioneer.
1921: A Czech writer Karel Capek used the word Robot for the first time in his play R.U.R.
1929: A Japanese professor made the first Japanese robot.
1925: Wilhelm Lenz and Ernst Ising created the first Artificial Recurrent Neural Network (RNN) model.
1943: Warren McCulloch and Walter Pitts founded the first ever model of Artificial Neurons. The inspiration was taken from the human brain and its functionality. This laid the foundation for the first mathematical description of ANNs.
1949: A Hebbian theory was introduced to explain ANNs.
1949: A book was released named “Machines That Think”. In this book, Edmund Callis Berkley explains the advancements made in computers then and their limitations.
Contributions Of John McCarthy
One of the most influential name in the development of Artificial Intelligence is John McCarthy. He worked as a foundation in the development of AI and discovered new technologies with his experiments. His early discoveries in AI paved the way for other scientists. Born on September 4, 1927, John McCarthy was an American mathematician and computer scientist. Throughout his career, he received many awards including the Turing Award in 1971, the National Medal of Science in 1990, and the Kyoto Prize.
McCarthy had worked at several institutions such as Dartmouth College, MIT, and Stanford University. He was the one to introduce the term Artificial Intelligence (AI). For which he is known as the Father of AI. He spent most of his career at Stanford University and died on October 24, 2011. John McCarthy has many contributions in the field of AI but here I am pointing out the most crucial ones.
1956: Introducing the Word “Artificial Intelligence”
In 1956, McCarthy conducted a Dartmouth Conference at Dartmouth College. This is where John McCarthy officially introduced the term Artificial Intelligence. From then AI was established as a field of study. At the Dartmouth Conference he defined AI as the “Science and Engineering of Making Intelligent Machines.”
1958: First AI Programming Language (LISP)
John McCarthy created a programming language named LISP in 1958 at MIT. LISP stands for “List Processing” and all operations and functions in LISP are written as lists with the operator or function name first and its arguments next. The LIST language became very popular for the development of AI because of its excellent support for symbolic reasoning and manipulation. LIST is still popular today.
1959: Developing Time-Sharing Concept
Time sharing concept allowed multiple users to access a computer system simultaneously by sharing processing time. This concept was used in the 1950s and 1960s when computers were large, expensive and could only be used by one person at a time. McCarthy’s work on time-sharing greatly improved the efficiency of operating systems. This concept paved the way for modern multitasking and shared computing resource systems.
1959: Concept of Garbage Collection
While developing the LISP language, McCarthy introduced the idea of garbage collection. The basic principle of this concept was automatically managing memory in computing systems. This concept allowed AI programs to manage memory more efficiently by freeing up unused memory spaces. Garbage collection became a standard feature in many languages after LISP and is still used today in many programming languages.
1980s: Logic-Based AI and Circumscription
McCarthy introduced the concept of circumscription, a method of non-monotonic reasoning. In traditional logic systems, once something is proven true it remains true forever. But in real world situations, knowledge is often incomplete and can change from time to time. Non-monotonic reasoning allows AI systems to make conclusions based on new information, which is important for handling uncertainty. If you are researching about AI then 10 cool facts about AI can be beneficial for you.
Contributions Of Alan Turing
Alan Mathison Turing born on June 23, 1912, was a British Mathematician. Turing made many contributions to the field of AI and invented the computer called Turing Machine. Turing is known as the father of theoretical computer science and artificial intelligence. He tragically died from cyanide poisoning on June 7, 1954.
1936: The Turing Machine
Alan Turing introduced the concept of Turing Machine in his 1936 paper “On Computable Numbers, with an Application to the Entscheidungsproblem.” The Universal Turing Machine is also considered a model for today’s modern computers. It is a theoretical model of performing calculations that defines “An abstract machine is capable of performing any algorithmic calculation”.
Turing Machine can defined by imaging an infinitely long tape with storage cells. The tape is covered with symbols that tells the machine how to manipulate with other symbols. Turing’s Machine was first named “ A-machine” by Alan Turing himself.
1948: Turing’s contribution to ANNs
Turing published a report named “Intelligent Machinery” in 1948. In his report, he discussed the idea of machines that could learn. Turing then introduced a machine “Unorganized Machine”, this was a theoretical machine. He thought that by mimicking how the human brain works through interconnected neurons, correspondingly machines could learn from experience and can improve themselves. This machine was an early model of ANNs.
1950: The Turing Test
In 1950, Turing introduced a test in his report “Computing Machinery and Intelligence.” In this test, a human evaluator and a machine interact with each other. The human interacts with the machine through a text only interface and must decide whether he is conversing with a human or a machine. If the Human cannot distinguish the machine from the human, the machine is considered to have passed the Turing Test.
1950: The Imitation Game
The Imitation Game is another variation of The Turing test. In this test, a human interacts with both a machine and another human via a computer interface. The judge human has to identify which is human and which is machine based on their responses. If the human cannot distinguish between the machine and the human, the machine is considered to have passed the test. The both tests were designed to answer the question “Can machines think?”
1951-1974: The Golden Phase of AI
This phase is known as the Golden phase of AI as this phase was full of discoveries and inventions in Artificial Intelligence. The important events of this phase are as below:
1951: Marvin Minsky and Dean Edmonds created the first Artificial Neural Network (ANN) and named it SNARC.
1952: A computer scientist, Arthur Samuel created a program that was designed to play checkers on its own.
1955: Allen Newell and Herbert A. Simon created the first AI program and named it Logic Theorist.
1956: A research conference was conducted at Dartmouth College. The US government funded this research and many researchers concluded that machines will be able to think like humans within a century.
1956: John McCarthy Introduced the word “Artificial Intelligence” for the first time at the Dartmouth Conference.
1958: John McCarthy developed a programming language named LISP at MIT.
1959: Arthur Samuel introduced the word “Machine Learning” for the first time.
1961: The first industrial robot “Unimate” was set to work at a production line in the automotive industry.
1964: Daniel Bobrow created an early program for Natural Language Processing (NLP) at MIT.
1966: Joseph Weizenbaum created the first ever AI chatbot named “ELIZA”. ELIZA is an early NLP program to see the communication between humans and machines.
1966: SRI (Stanford Research Institute) the first mobile robot “Shakey the Robot” in this year. The was the first robot that was able to do logical reasoning and can perform physical actions.
1969: An expert system “DENDRAL” was developed to identify organic molecules.
1970: The early Expert System “MYCIN” was developed at Stanford University. This expert system was designed to diagnose bacterial infections.
1972: A logic programming language “Prolog” was developed University of Marseille. This language later became popular in AI research.
1973: The UK government commissioned a report (Lighthill Report) on AI research. This report criticized the lack of progress in Artificial Intelligence and then the UK government reduced the funding for AI research. This reduced budget for AI projects ultimately led to the First AI Winters.
1974-1980: AI winters (First Phase)
The period from 1974 to 1980 is often referred to as AI winters. This phase showed a decline in the development of AI. During this phase, the government constantly decreased funding for AI research. Because of this the interest of the public in AI started disappearing slowly. The important event of this phase is below:
1979: The American Association of Artificial Intelligence (AAAI) was founded in this year. It was then renamed the Association for the Advancement of Artificial Intelligence (AAAI).
1980-1987: Boom of AI
1980: The first conference of the Association for the Advancement of Artificial Intelligence (AAAI) was conducted at Stanford.
1981: An expert system “XCON” was developed by Carnegie Mellon University. It was used by a corporation to configure its computers.
1982: The Japanese government allocated $850 for a Fifth Generation Computer Systems project (FGCS). The main purpose was to create the Fifth-generation computer.
1985: This year an artist Harold Cohen created a program that was capable of creating original artistic images. This program was named “AARON and was demonstrated at the National AAAI Conference.
1986: Ernst Dickmann along with his team at Bundeswehr University created the world’s first driverless car. The top speed of this driverless car was 55mph.
1987: A managerial advisory system Alacrity was launched by Alctrious Inc. This system used a complex expert system with 2000+ rules.
1987-1993: AI winters (Second Phase)
The so-called second phase of AI winters begins now. Decreased funding and less interest of people was also the reasons for these winters. The AAAI had already warned about it in 1984. Some important dates are:
1987: The specialized LISP-based hardware collapsed this year and because of this many specialized LISP companies were failed.
1988: A chatbot named “Jabberwacky” was invented by Rollo Carpenter.
1993-2011: The Rapid Growth of AI Agents
1997: AI program “Deep Blue” built by IBM defeated a champion Garry Kasparov in chess. This was a remarkable event in AI’s history where an AI beats human intelligence.
1999: Sony introduced a robotic dog “AIBO”. AIBO was the first Artificial Intelligence pet.
2000: This year Dr. Cynthia Breazeal made a robot “Kismet” as an experiment in affecting computing. The robot was able to recognize and interact with human emotions.
2000: Many companies started using Interactive Voice Systems (IVR) to improve their customer service.
2002: A vacuum cleaner robot “Roomba” was introduced to do household chores. These types of AI help us in our daily lives highlighting AI benefits to us.
2004: NASA sent two robotic rovers Spirit and Opportunity to Mars. We were able to explore Mars with Artificial Intelligence without physical presence of humans.
2006: James Moor organized a conference “AI@50” in the remembrance of Darth Mouth Conference. This was the 50th anniversary of the Darth Mouth Conference held in 1956.
2006: Amazon launched its virtual assistant “Alexa”. Introducing virtual assistants revolutionizing the way of interacting with technology.
2010: Microsoft launched Xbox 360 Kinect which can track human body movement with the help of AI. The purpose was to allow users to play Xbox wirelessly.
2011: Apple launched the popular virtual assistant “Siri”. Also, many companies started to launch their Virtual Assistants after this.
2012 to Onwards: Modern Era of AI
This phase is known as the phase of AGI (Artificial General Intelligence) or generative AI because of the huge surge in developments of AI.
2015: Two new techniques 1. Highway Network and 2. Residual Neural Network were introduced to train deep neural networks.
2015: An open letter was signed to stop and ban the use of weapons in AI and robotics.
2016: The most famous humanoid female robot “Sophia” was introduced. Sophia was developed by Hanson Robotics in Hong Kong.
2018: Google launched its powerful NLP model “BERT” to improve the accuracy of its search engine.
2019: Google DeepMind launched “AlphaStar” which was designed to play StarCraft II (A video game). This software out-beaten all human players and achieved the grandmaster status.
2021: OpenAI launched an AI model Dall-E which was capable of generating images from text prompts.
2022: OpenAI launched the popular chatbot “Chat GPT-3.5” transforming the industry of content generation. Chat GPT 3.5 became very popular in no time attracting 100 million users in the first 2 months.
2023: OpenAI launched the Chat GPT-4 model. This model was capable of generating images.
2023: Google launched its chat bot named “Bard”.
2024: OpenAI launched “SORA” which was capable of generating videos up to 1 minute long with text prompts.
2024: Meta launched their Large Language Model (LLM) Llama 3. Meta claims that this is the most capable openly available LLM.
Frequently Asked Questions
Where is the birthplace of AI?
The conference of Dartmouth held at Dartmouth College in 1956 is considered as the birthplace of AI.
When did AI become popular?
Artificial Intelligence became very popular in 1993-2011 because of huge developments and rapid invention of AI agents in this phase.
When was the first AI robot made?
The first AI robot “Shakey the Robot” was made in between 1966 and 1972 at SRI (Stanford Research Institute).
0 Comments