Part 1: What is Artificial Intelligence and where did it all begin?
As soon as computers came into our world, scientists began to think on how they would revolutionize our world and the impact they would make to how we live. Even as far back as Aristotle introducing the idea of associationism in 300BC and this would trigger our attempts to study the human brain.
Turing’s Puzzle Piece
By the 1950s there was a generation of mathematicians, scientists and philosophers investigating the concept of AI and the possibility of a computer being able to think for itself. One such man was Alan Turing, a young British Polymath who spent his time exploring the mathematical possibility that Artificial Intelligence could be a possibility. He believed that since humans are able to use available information as well as reasoning and sound judgement in order to solve a problem, then certainly machines could be able to do the same.
He built upon these ideas and they built a basis for his 1950 paper, Computing Machinery and Intelligence – in which he discussed the idea on building intelligent machines and suggested ways in which we could test their intelligence.
However, big challenges stood in his way; back then (1949) computers could only execute commands but could not store these commands, in essence they could do something but couldn’t remember what they did.
Computing was also extremely costly and very few were available to be leased. Leasing a computer run up to $200,000 a month. Only big firms and prestigious universities could afford day in day out usage of these big bulky machines.
Aside from all that for people to invest in the idea of machine intelligence, scientists had to go to extreme miles to prove that it really was worth pursuing.
Then came a conference that probably changed that whole perspective.
Founding Fathers of AI?
In 1954, Allen Newell, Cliff Shaw and Herbert Simon developed the Logic Theorist. A program that many believed was truly the first artificially intelligent program. It was designed to mimic the problem-solving skills of the human brain and was funded by Research and Development Corporation (RAND). This program then went on to be presented at the Dartmouth Summer Research Project on Artificial Intelligence. (DSRPAI) that was hosted and run by John McCarthy and Marvin Minsky in 1956. This conference brought together top researcher and minds from various fields for an open-ended discussion on Artificial Intelligence – a term coined by McCarthy during the event. Though the event fell short of their expectations. Everyone agreed that Artificial Intelligence was achievable, and this fueled the research undertaken for the next 20 years.
“from three to eight years we will have a machine with the general intelligence of an average human being.”
Marvin Minsky during an interview with Life Magazine
The Leaps of Faith
And a few turbulent winds
Well, from 1957-1974 AI took off and began to flourish, machines could store more data and computers became faster, cheaper and more accessible. Machine Learning algorithms began to improve and scientists in turn got better at knowing what algorithms to apply to their problems. These successes, majorly from the researchers who attended the conference, prompted investment form government agencies such as the Defense Advance Research Projects Agency (DARPA) to fund AI. They were particularly interested in the ability of a machine being able to transcribe and translate spoken language as well as big data.
People remained optimistic and so did Minsky, in 1970 Marvin Minsky told Life Magazine, “from three to eight years we will have a machine with the general intelligence of an average human being.” This was true but there was still a long way to go.
Once the initial cloud of dust surrounding AI pursuits had died down, a big problem emerged, the lack of computational power to make a huge enough leap. Computers could not store large enough information or process it fast enough. One of McCarthy’s students, Hans Moravec stated that “computers were still millions of times too weak to exhibit intelligence.” And slowly as patience from the investors and the industry began to die down, so did research and funding.
However something remarkable happened in the next decade.
Re-Kindled
In the 1980’s, AI was brought back to life mainly by the boost of funds and the expansion of the algorithmic toolkit. John Hopfield and David Rumelhart came along and popularized the term “deep learning” a technique in which a computer learns through experience. Then came Edward Feigenbaum who introduced expert systems which helped in industry development and are still widely used today.
$400 million was invested by the Japanese Government between 1982-1990 for the development of expert systems and other related endeavors under their Fifth Generation Computer Project (FGCP) initiative.
AI thrived and through the 1990s to the 2000s many of the landmarks and targeted goals had been achieved.
In 1997, reigning world chess champion and grand master Gary Kasparov was defeated by IBM’s Deep Blue, a chess playing computer program. This highly publicized match was the first time a reigning world chess champion loss to a computer and served as a huge step towards an artificially intelligent decision making program.
The History of Artificial Intelligence by Rockwell Anyoha
Fast Forward >>>
AI is not recent, it is instead built upon the shoulders of giants
Today we haven’t radically changed the way we compute and code artificial intelligence. It turns out that the very storage limit that computers faced over 30 years ago is no longer a problem. As Moore’s Law states, memory and the speed of computers should double every year. That way computers have been able to catch up and surpass our needs. Deep Blue beating Gary Kasparov in 1997 and Google’s Alpha Go being able to defeat Chinese Go champion, Ke Jie.
Through the immense roller coaster, AI research has been able to grow and we hope that in the next few years it can keep up with humanity as development continues to occur.
What is Artificial Intelligence?
Well it depends on who you ask.
The fathers of the field Minsky and McCarthy, described Artificial Intelligence as any task performed by a program or a machine that, if a human carried out the same activity, we would say the human had to apply intelligence to accomplish the task.
Pretty broad but this is why there are many arguments and discussions on whether something is truly AI or not.
AI systems will typically demonstrate at least some of the following behaviours associated with human intelligence: planning, learning, reasoning, problem solving, knowledge representation, perception, motion, and manipulation and, to a lesser extent, social intelligence and creativity.
Today we can thank these brave souls for their immense contribution to the way we live today and the future ahead:
Marvin Minsky, Ray Solomonoff, Claude Shannon and John McCarthy amongst others.
Resources
A major part of this publication is adapted from: The History of Artificial Intelligence by Rockwell Anyoha
Dartmouth Summer Research Project on Artificial Intelligence
Sparky Science 