How finance organizations deal with AI/digital transformation challenges

Shahzad Mahmood is a notable financial professional with over twenty years of broad finance experience currently serving as IBEX Philippines’s CFO and the Head of Global Treasury NASDAQ(IBEX). Shahzad’s dual position at IBEX affords him a unique global economics, treasury, and finance perspective. He has also previously held positions as finance head in Western Africa and CALA regions for IBEX and held various managerial positions in subsidiaries of Axiata, Grant Thornton, and Deloitte.

In a conversation with Prisila, a correspondent in Asia Business Outlook Magazine. Shahzad discussed AI's influence on finance, exploring challenges in traditional processes, assessing AI's impact on algorithmic trading, and addressing fairness and stability concerns in financial institutions.

How has AI impacted traditional financial processes, and what are the key challenges that arise from this trans- formation?

Artificial intelligence (AI) is the advanced stage of digital transformation. Digital transformation has facilitated the automation of basic tasks, resulting in a notable decrease in human error and enhanced operational effectiveness. However, the integration of artificial intelligence (AI) has empowered the finance sector to leverage predictive analytics. This capability enables the identification of irregularities in transactions to detect fraudulent activities, ultimately mitigating risks for financial organizations.

While this advancement allows for the transformation of financial operations, it also presents new issues, with data privacy and security emerging as the primary concern. The sensitive nature of financial information necessitates robust safeguards to prevent unauthorized access. The majority, if not all, of artificial intelligence (AI) models rely on extensive datasets and substantial computational resources. Consequently, restricted-access information may end up being used in an algorithm not originally intended and stored in an unauthorized location. Therefore, security and accountability are essential components of AI models and decision processes, particularly in financial operations.

How does digital transformation impact customer experience in the financial sector, and what challenges might arise in maintaining a balance between innovation and security?

The digitalization of the financial sector yields favorable outcomes, such as heightened convenience, tailored services, and greater operational effectiveness. Nevertheless, innovation is invariably accompanied by a set of obstacles, including but not limited to cybersecurity vulnerabilities, adherence to legal requirements, and the incorporation of existing legacy systems. Therefore, it is vital to promptly and meticulously tackle these issues in order to uphold a harmonious equilibrium between the advancement of novel ideas and the preservation of safety and protection.

As part of the  Business Process Outsourcing (BPO) sector, with the primary objective of consistently delivering exceptional client experiences, I have long held the conviction that prioritizing innovation is a prerequisite to bringing about change. The change will certainly lead to valid security concerns.  The proliferation of digitization and the advent of artificial intelligence has facilitated the emergence of heightened and intricate cybersecurity vulnerabilities and incursions. The sooner we address those security concerns, the better.

How can financial institutions address the ethical concerns associated with using AI in decision-making processes?

The very first step in addressing ethical considerations related to the utilization of AI decision-making involves ensuring transparency and effective communication regarding the functioning of AI algorithms, the decision-making process itself, the specific factors involved, and their corresponding significance within the AI model.

One of the major obstacles to achieving openness in the field of artificial intelligence (AI) lies in the proprietary nature of AI algorithms that inherently possess commercial value. The resolution of this issue necessitates a predetermined set of regulations for the operation of AI systems while giving consideration to ethical problems. Addressing these ethical concerns can be accomplished by prioritizing openness, adopting ethical AI principles, and implementing frequent audits. Addressing the issues would alleviate customer concerns and uphold their rights and privacy.

Examine the impact of AI on algorithmic trading in financial markets. What challenges do financial institutions face in ensuring the fairness and stability of algorithmic trading systems?

Algorithmic trading in financial markets has been found to enhance efficiency as a result of the utilization of artificial intelligence (AI) algorithms. These algorithms provide faster transaction execution compared to human capabilities, hence contributing to the overall efficiency of financial markets; however, they potentially exacerbate short-term market volatility, particularly in unforeseen events or extreme market conditions.

The consideration of fairness issues is of utmost importance, given that artificial intelligence (AI) models have the potential to perpetuate biases inadvertently. Consequently, it becomes imperative for institutions to assume responsibility in addressing these concerns.

Operational hazards, such as technical malfunctions, present intricate challenges, underscoring the importance of comprehensive risk management.

"One of the major obstacles to achieving openness in the field of artificial intelligence (AI) lies in the proprietary nature of AI algorithms that inherently possess commercial value"

In what ways can AI be used for financial forecasting and planning, and what challenges may organizations encounter in relying on predictive analytics for strategic decision-making?

Artificial intelligence (AI) enables the implementation of dynamic scenario analysis, which empowers enterprises to create models and conduct simulations of diverse financial situations in response to fluctuating market conditions. This functionality facilitates the development of financial plans that are more adaptable and resilient.

In preparing all financial forecasts and planning, the primary objective is to ensure that the forecasts represent the best prediction of the future outcome. With that in mind, to enhance the efficacy of AI models in financial forecasting and planning, it is imperative for the model to utilize information beyond customary historical data and apply fundamental regression techniques.

This can only be achieved by developing and continuously refining sophisticated AI models incorporating a wide range of datasets and a range of potential decisions under each of the expected outcomes with the aim of achieving the objective of being the best prediction of the future outcome. The model continues to learn from actual events and dynamically adjusts its future actions.

Organizations increasingly prioritize using predictive models and are inclined towards potentially diminishing the reliance on human expertise and intuition. However, the excessive dependence on these models may pose a potential risk if they cannot incorporate unanticipated circumstances or abrupt shifts in the market.