QU’s College of Law publishes article discussing benefits and risks of AI in financial institutions

THE Center for Law and Development at atar University’s College of Law recently published an article under the title ‘Banking on AI: Mandating a Proactive Approach to AI regulation in the Financial Sector’.

In the article, Dr Jon Truby, Dr Rafael Brown and Dr Andrew Dahdal from the Center for Law and Development argue that the best way to encourage a sustainable future in AI innovation in the financial sector is to support a proactive regulatory approach prior to any financial harm occurring.

The article, which was published in Taylor Francis’ Financial Markets Review on May 15, 2020, is part of an NPRP grant (11S-1119-170016) from the atar National Research Fund ( NRF).

The authors discuss the benefits and risks associated with integrating artificial intelligence (AI) into the existing systems and processes of financial institutions. AI for financial institutions offer many benefits such as enhanced fraud detection, more accurate lending and credit assessments, stronger cybersecurity detection and faster regulatory compliance.

However, the authors also emphasise the need to consider AI’s inherent risks and threats. They argue that a proactive approach should implement rational regulations that embody jurisdiction-specific rules, in line with carefully construed international principles.

The authors use a riskbenefit analysis framework to examine three distinct contexts in which AI can be utilised: (1) how financial service providers use AI in relation to their clients, (2) how financial services firms use AI in their compliance efforts, and (3) how regulators use and may use AI in their regulatory efforts.

According to the authors, financial intuitions are introducing a plethora of AI-driven financial services, including robo-advising, algorithmic investing and insurance/ credit assessment at a rapid pace and as a first-generation technology. Though governed by existing financial and data protection laws, neither the developers nor the financial institutions have significant legal obligations in any jurisdiction to follow the international principles on AI governance, which in turn have been developed to require accountability, transparency and fairness in the utilisation of AI software in the financial sphere.

The authors contemplate it to be prudent and timely for regulators to seriously consider the nature and scope of AI regulation in the financial services sector. They argue that the adoption of rational regulations that encourage innovation whilst ensuring adherence to international principles will significantly reduce the likelihood of development of AI-related risks into systemic problems.

To accomplish it efficiently, the article suggests that policymakers intervene early with targeted, proactive but balanced regulatory approaches to AI technology in the financial sector that are consistent with emerging internationally accepted principles on AI governance. To explain attempts to attain such balance, the article provides the analysis of AI regulatory policies in different jurisdictions such as China, India, EU and Japan.

Considering that the AI regulation has significant geopolitical ramifications, it is contended that it will trigger race for global economic supremacy and will face strong opposition given the stakes at play. Therefore, it is asserted that balancing the risks of AI with the benefits of innovation requires addressing macro and micro level details. Taking account of this, it is proposed that policy makers in individual jurisdictions acknowledge the potential global significance of the sector in formulating their regulatory frameworks.

The article also emphasises the challenge that policymakers will face in balancing innovation with potential risks to the public good due to the AI development. In this regard, it puts forwards the argument that any attempt at manipulating the future regulation of AI to prefer one stakeholder to others will likely result in damaging market distortions in the financial sector.

Having provided thorough analysis, the article concludes that in order to avoid “distorted” and “unbalanced” outcomes, AI regulation in the financial sector must be implemented in a consistent and balanced manner across all three contexts by taking into consideration the interests of all stakeholders.

The article is available at https://doi.org/10.1080/1752 1440.2020.1760454.