Beyond the hype: How can the financial industry benefit from artificial intelligence?

Abstract

Artificial intelligence (AI) has been in the spotlight for several years now and even remains so during the difficult COVID-19 period. There has been a lot of AI hype and a lot of AI-related ‘buzz-phrases’ such as ‘data is the new oil’, fueled by our imagination. In practice, nowadays, AI is more related to machine learning (ML). In this paper, we will review the different approaches that defines ML to have a better sense of the concrete strengths and weaknesses of each. In particular, we will tackle some misconceptions, which involve some mismatches between expectations and practical results. In addition, we will also see how some breakthroughs from academics, such as beating the Go world champion, could translate in applications for the financial industry. Finally, we will see the implications for business and organisation, beyond technical aspects. Indeed, the AI hype tends to create a lot of noise, making harder for decision makers and executives to distinguish noise from actionable signals. For instance, change of organisation culture is perhaps the most important for successfully deploying ML systems in real use cases. It requires changing many things from recognising the value from data from existing processes; increasing the organisation’s risk appetite; creating new teams and fostering internal and external collaborations. As an example, we will review the strategy from Quebec’s Autorité des Marchés Financiers (AMF) for implementing these changes. Beyond a specific organisation, the whole financial industry could benefit from this new wave of innovative technologies as other sectors.