Artificial Intelligence (AI) may have only recently come to the fore in the last ten years, but did you know that the quest for AI began over hundred years ago?
This is the timeline as we see it, of major events in AI. We will add to it as we get suggestions of major developments occur.
Think we are missing something major? Get in touch >> https://www.ethics.ai/contact/
AlphaGo Zero – DeepMind
DeepMind technologies was acquired by Google in 2014. However it wasn’t until 2016 that the technology was put into the spotlight when the AlphaGo programme – which is built using deep neural networks – beat the European Go Champion – Fan Hu. This was the first time an AI defeated a professional Go player.
Apple’s Siri (2011), Google’s Google Now (2012) and Microsoft’s Cortana (2014) are the first phone apps to use NLP to answer questions, perform actions and make recommendations.
MAN VS. MACHINE ROUND 2
IBM had already won one battle with Deep Blue but fancied taking on the human brain once again, this time with their AI platform – Watson.
Waston took on the US quiz show, Jeopardy, where it had to answer riddles and complex questions. IBM trained the system for over three years and used a number of AI techniques including neural networks in order for the system to understand patterns in questions and answers.
Watson demolished its opposition – the victory went viral and was hailed as a big win for the AI industry.
Despite millions in investment and thousands of hours of research, speech recognition had never reached an accuracy that made it viable as a commercial solution. That was until 2008 when Google pioneered a new AI approach using distributed networks to learn and spot patterns in the vast volumes of data captured from Google’s customer base.
The technology was deployed into the Google app on the Apple iPhone and boasted accuracy of over 92%.
The US military invested heavily in robotics and AI in 2005. BigDog built by Boston Dynamics was one of the first autonomous robots to be built for war. Despite being built to work in terrain too rough for conventional vehicles, the robot never actually saw active service.
iRobot also had some skin the game with their bomb disposal robot, PackBot, which ended up being deployed in Iraq and Afghanistan.
Roomba manufactured by iRobot was the first commercially successful robot. Although a far cry from the lofty ambitions of early AI pioneers, Roomba had enough intelligence to reliably and efficiently hoover a home.
Roomba ushered in a new era of autonomous robots.
MAN VS. MACHINE
In 1997 IBM’s supercomputer – Deep Blue – competed with world chess champion – Garry Kasparov. Deep Blue was capable of evaluating 200 million positions a second and convincingly beat the former champion. Some say this is the moment that AI came of age. But for others, this was just a demonstration of brute force at work on a highly specialist problem with clear rules.
POLLY: BEHAVIOUR-BASED ROBOTICS
Polly was the first mobile robot to move at animal-like speeds (1m per second) using computer vision for its navigation.
A DECADE LATER & A COMMERCIAL SOLUTION
Ten years on from the AI winter and the world of intelligent machines raised its head once again – this was when AI’s commercial value started to be realised and investment started to get pumped back into the industry.
The new commercial systems were less ambitious than the early days of AI. Instead of trying to create a general intelligence system – they focused on much narrower, specialist tasks.
The first successful commercial system, known as RI, was built by the Digital Equipment Corporation to improve the process of configuring orders for new computer systems.
It was reported that in 1986, the company was saving over $40m a year thanks to RI.
KNOWLEDGE BASED MEDICAL DIAGNOSIS PROGRAM
In hospital tests, the computer program was occasionally put up against the intuition of nationally respected doctors as each sought to diagnose a complex case. At times, the program sorted through the symptoms and came up with the correct answer when the doctors could not.
AI FUNDING DECLINES
By the early 70’s the AI industry was in trouble. Millions had been spent on research and development and the only thing to show for it was Shakey!
There was strong criticism from US Congress and in 1973, a leading mathematician gave a damning report on the state of AI. His view was that AI machines were not even up to the challenge of supposedly simple tasks.
Funding for the industry was subsequently slashed, ushering in what became known as the AI winter.
THE FIRST ‘INTELLIGENT’ ROBOT LAUNCHES IN THE UK
AI was certainly lagging behind the predictions made by Minsky et al. which was quickly made apparent by the launch of Shakey the robot.
Researchers spent six years developing Shakey – which was the first ever robot to make decisions about its actions by analysing its surroundings. However, Shakey was met with a cold reception – the robot was painfully slow and in some cases would stop and wait for an hour whilst it planned its next move. It was back to the drawing board…
2001: A SPACE ODYSSEY
Just like the books of the 50s – AI was now creeping into the cinema. Minsky advised on the Stanley Kubrick film 2001: A Space Odyssey which featured an AI machine called Hal 9000.
In one scene, a mission scientist explains that he believes that HAL may well have genuine emotions. The film mirrored some predictions made by AI academia at the time that machines were heading towards human level intelligence very soon.
Mac Hack VI
In the spring of 1967, Mac Hack VI played in the Boston Amateur championship winning 2 games and drawing 2 games. Mac Hack VI beat a 1510 United States Chess Federation player. This is the first time a computer won a game in a human tournament. At the end of 1968, Mac Hack VI achieved a rating of 1529. The average rating in the USCF was near 1500.
ELIZA is an early natural language processing computer program created from 1964 to 1966 at the MIT Artificial Intelligence Laboratory by Joseph Weizenbaum. Created to demonstrate the superficiality of communication between man and machine, Eliza simulated conversation by using a ‘pattern matching’ and substitution methodology that gave users an illusion of understanding on the part of the program, but had no built in framework for contextualising events.
The first industrial robot, Unimate, joined the assembly line at a General Motors plant to work with heated die-casting machines.
GENERAL PROBLEM SOLVER
The General Problem Solver (GPS) was the first useful AI program, written by Simon, Shaw, and Newell in 1959. GPS was designed to solve lots of kinds of problems, using “reasoning” mechanism.
AI PROGRAMMING LANGUAGE
John McCarthy develops Lisp – which quickly becomes the most popular programming language used in AI research.
You may not have heard of him, but John McCarthy was the first person to coin the term ‘Artificial Intelligence’, which he defined in a publication that he launched at a summer conference at Dartmouth University in the summer of 1956.
At the conference, a number of top scientists discussed how to tackle AI. Some of the more influential academics, like Marvin Minsky, favours a more top-down approach which is to pre-programme a computer with a set of rules that govern human behaviours.
Others preferred the bottom-up approach such as neural networks that act like brain cells and learned new behaviours.
Over time Minsky’s views dominated, and together with McCarthy – a computer scientist – they won substantial funding from the US government, who hoped AI might give them the upper hand in the Cold War.
SCI-FI BOOKS PAVE THE WAY
Isaac Asimov was a true visionary and in 1950 published ‘I Robot’ – a collection of short stories about intelligent robots.
Asimov was one of several sci-fi writers who picked up on artificial intelligence and imagined how the future could look. His work was popular and helped to inspire a generation of scientists. Remarkably a lot of what Asimov predicted has rung true – such as his idea that in the future computers would be capable of storing all human knowledge that anyone can ask any question.
Asimov is probably best known for his book ‘Three Laws of Robotics’ – which is designed to stop our AI creations from turning on us. Something which is still at the forefront of the minds of most in the industry.
The Turing test, developed by Alan Turing in 1950, is a test of a machine’s ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human.
World WAR 2 TRIGGERS NEW SCHOOL OF THOUGHT
WW2 brought together some of the country’s leading scientists across a multitude of disciplines including computing and neuroscience.
In Britain, neurologist Grey Walter and mathematician Alan Turing were two of the brightest minds in their respective fields and were tasked with tackling the challenge of intelligent machines.
They traded ideas during the war with Walter building some of the first ever robots. Turing went on to invent the Turing Test, which is a method for determining whether a machine passes as a human or not and can truly be called AI.
Programmable mechanical calculating machine, written by Ada Lovelace