The term artificial intelligence was coined at a Dartmouth University workshop in 1956. Two participants took issue with it, however. For years, they insisted instead on the terminology of complex information processing, a less evocative but more exacting description of the discipline, which includes statistics, computational science and machine learning.
The connection between AI and financial services goes back to 19th century computing pioneer Charles Babbage, who viewed London’s Bankers’ Clearing House as an information processing system. Yet the work of English mathematician Alan Turing almost a century later first prompted academics to believe that generalised computer intelligence could be achieved.
In the 1980s, an entrepreneurial group of new investment managers began to consider applications of AI. Renaissance Technologies and D.E. Shaw, employing techniques from statistics and computer science, launched in the United States. In London, the firm Adam, Harding & Lueck Limited, was pioneering the application of computer simulation to systematically trade futures markets. These firms and their progenies – including Winton Group and Two Sigma Investments – are today among the world’s most successful quantitative investment firms.
By the 1990s, fund managers including Fidelity and LBS Capital Management were trying – and failing – to use neural networks, a type of machine learning, to identify investments. Nonetheless, a growing number of institutional investors used statistical and computer science techniques throughout the 1990s to amass data, identify trends, and trade markets.
The current surge of interest in AI has again centred on neural networks, thanks partly to systems developed by Alphabet subsidiary DeepMind to play the Chinese game of Go. But games like Go or chess are what statisticians term “fully observable” – they have defined and constant rules, and a large yet finite number of potential permutations. By contrast, global financial markets are human institutions with ever-changing characteristics. They present computers with a far tougher challenge.
While caution is thus warranted, investment managers should continue to gain from AI – as they have over the past 30 years. There has been substantial growth in computing power and memory – products of micro-processing efficiency gains described by Moore’s Law. Advances in automatic data capture similarly hold out promise.