By Ernest Amoabeng Ortsin*
The moment I received news about my selection to participate in the Leading with Artificial Intelligence (AI) Lab under the auspices of Global Leadership Academy (GLAC), International Training Centre of the ILO (ITCILO) and FAIR Forward, I told myself that I needed to research. As a newbie in the field of AI, I knew I needed to acquaint myself with some of the rudimentary information on the subject.
Throughout my research, especially when I wanted to get an understanding of the evolution of AI, I came across several names. These included Charles Babbage, Ada Lovelace, Alan Turing, John McCarthy, Claude Shannon, Nathaniel Rochester, and Marvin Minsky. I devoted some time to read about these personalities and the roles they played in advancing the frontiers of AI.
Today, I am going to reflect on one of them, Alan Turing, who appears to have been the first to make a rather forceful argument about the possibility of machines being capable of thinking. Turing was a British computer scientist born in 1912. After studying at the King’s College (University of Cambridge) from 1931 to 1934, he proceeded to Princeton University where he specialised in mathematics and cryptology, earning a Ph.D. in 1938.
During the Second World War (1939 – 1945), Turing, who had by then returned to Cambridge, worked with a team of codebreakers at the Government Code and Cypher School (GC&CS) located at Bletchley Park. It was the responsibility of Turing and his team to crack or decipher coded communications that the Allies intercepted from the German troops. Without doubt, Turing and his team were critical in the defeat of the German troops as the Allies were able to foil their tactical plans.
After the war, Turing turned his attention to the then burgeoning field of computer science. In 1949, he romanticised the idea of creating a ‘brain’. That is, he thought of making machines that could think like humans. This spurred him to publish a rather thought-provoking journal article titled ‘Computing Machines and Intelligence’ in 1950.
Quite enigmatically, Turing began his paper by stating, ‘I propose to consider the question, “Can machines think?”’ The paper is about 28 pages and Turing was highly animated in his arguments that it is possible for machines to mimic human thinking.
A careful reading of the paper shows that Turing may not necessarily have been the progenitor of the idea that machines can think. It seems the subject of thinking machines was highly topical around the decade of 1940, and he belonged to the school of thought that believed that it was possible for machines to think. This is quite evident in the argumentative manner he presented his paper.
As a matter of fact, Turing devoted a substantial amount of time and energy to refuting what he described as arguments and objections against the idea of thinking machines. The arguments and objections were nine in number, and they notably included the idea that, theologically, humans have the ability to think because they have souls in them. And, that, in the absence of a soul, a machine cannot think.
The other arguments and objections related to the fact that, physiologically, humans are capable of thinking because they have a body that is imbued with a nervous system. Additionally, the other sets of arguments and objections raised the issue of human consciousness and extra sensory perception (ESP) that make humans unique in their cognition.
Turing argued that, in spite of all the doubts, he was optimistic that in 50 years (i.e. year 2000) there would be computers capable of thinking. He proposed a test by which machines can be qualified as having acquired thinking capabilities. In later years, it became known as the ‘Turing Test’.
According to the Turing Test, if within a period of five minutes, a machine can give written responses to written questions in a manner that convinces the questioner, at least 30 percent of the time, that it is not a machine but human, then that machine would have acquired thinking capabilities.
For several years the Turing test remained unsurmountable until 2014 when a Russian chatbot christened Eugene Goostman was adjudged to have finally passed the test in a competition that was organised at the University of Reading to commemorate the 63rd anniversary of the death of Turing. Quite intriguingly, the chatbot was first developed in the year 2001, perhaps fulfilling the vision Turing saw in 1950.
In the concluding part of his paper, Turing presented what indisputably passes as a blueprint for machine learning. He proposed that if a child’s brain could be studied to establish how the child, through a process of education, acquires adult thinking capabilities, the same process could be adopted to train machines on how to think like humans.
Indeed, there is no doubt that modern algorithms in machine learning, natural language processing, and the entire spectrum of AI have benefited immensely from the foundational theoretical thoughts of Turing. It is quite unfortunate that he did not live long enough to see his test passed as he died tragically in 1954 at the age of 42. His early demise notwithstanding, his memory lives on as a visionary and trailblazer in the field of AI.
* Ernest Amoabeng Ortsin is an Africa-based researcher with a growing interest in Artificial Intelligence policy research. He studied Political Science at the University of Ghana. He was a participant in the Leading with AI Lab and is a co-founding member of Leading with AI blog.
First Event of the AI Leadership Academy
TODAY - 18th February 2021 @ 15.00 CET (Brussels)
Equality in the digital era:
AI and anti-discrimination law in Europe
The AI Leadership Academy is an initiative aiming to inspire articles from different topics. We want to spread the knowledge that enables different stakeholders from different sectors to play a role in leading sustainable development with AI.
With the aim of achieving this goal, the Leading with AI blog founders have invited Raphaële Xenidis to share her experience in the field of AI and anti-discrimination law in Europe.
Speaker Biography: Raphaële Xenidis is currently a Marie Curie Fellow at courts at the Faculty of Law of the University of Copenhagen and a lecturer in European Union Law at the University of Edinburgh Law School. Previously, Raphaële was a researcher in equality law at the International and European Law Department of Utrecht University, where she worked for the coordination team of the European Network of legal experts in gender equality and non-discrimination, and as a researcher with the RENFORCE research centre, the Netherlands Institute of Human Rights (SIM) and the Gender and Diversity Hub. Raphaële obtained her Ph.D. at the Department of Law of the European University Institute where she wrote her dissertation on the question of intersectional and multiple discrimination in EU and ECHR law.
This free, virtual event will take place TODAY, Thursday, 18th February at 15.00 CET (Brussels). Please register via this Eventbrite link.
We look forward to seeing you there to share the knowledge to lead with AI