The Growing Impact of Artificial Intelligence on Financial Institutions

The initial phase of the fintech revolution was the digitization of processes and procedures. The
next phase, which is already underway, is the cognification of that digital infrastructure.
Financial Institutions are positioned to experience an enormous amount of disruption as a result
of the corresponding rise of Artificial Intelligence. New risks will emerge that must be managed,
further dramatic change will occur in how FIs function, and additional consolidation will likely
occur in the industry. This will be prompted by the adoption of AI-driven technologies and the
desire to acquire more data. In the future, the acquisition and manipulation of data will transform
how FIs (and other organizations) determine their own value.

This may ultimately result in creation of a new accounting standard - the Enterprise Value of
Data – which could become an integral part of financial statements, capturing the value of the
largest and most ignored corporate asset: data. Meaningfully defining and incorporating EvD on
to balance sheets will help ensure that corporate accounting and risk management standards
incorporate an AI-driven future. It will also put firms that are not specifically data-oriented on an
operational equilibrium with those that are. However, those FIs that fail to embrace their own AI
future will become less relevant, competitive, and profitable in time.

As fintechs build digital platforms that can accumulate data in a way that is conducive to
implementing AI, traditional FIs are busy trying to keep their aging technology systems from
crashing. Hard as it may be to believe, some FIs are continuing to use 1980s-era computer
technology and are finding it difficult and costly to transition to the 21 st century. Those FIs
should, at a minimum, build more data lakes (large repositories that can hold structured and
unstructured data) as an interim measure, but soon enough, they will have fallen so far behind
the leaders in the FI industry that there will be no hope that they will ever catch up. That is
probably already the case.

While massive amounts of data are one key prerequisite for FIs to become viable contenders in
the race of AI supremacy, the data must be in a location where it can be processed according to
the rules of Big Data. FIs may therefore need to rely on external sources to harness the power
of Big Data through parallel computing. Some FIs may opt to implement advanced software
within their own data centers and manage the parallel computing process themselves. Others
may choose to leverage cloud providers to scale up their computing capacity, which raises
questions such as, what data can be sent outside of the organization and what data cannot, and
must the data first be anonymized? FIs will be grappling with such questions for many years to
come.

The increasing digitization of banks is a double-edged sword, however, which needs to be
handled with care. On one hand, FIs must digitize all of their operations as a matter of survival
in the new age of the digital economy because consumers demand it. Yet, when an FI digitizes
its end-to-end operations it opens itself up to an even greater extent to new and fast-moving
fraud and other forms of risk. Examples of instances when FI digitization has led to fraud or
systemic failure abound. Among the better known are the trillion-dollar 2010 flash crash caused
by a London-based trader who was spoofing, and the largest ever intra-day trading crash in the
US in 2018, resulting in a 1,600-point drop in a matter of seconds, and believed to have been
caused by programmed trading. This is why AI should be embedded in customer-facing
applications, but also in the back-end, to protect against fraudulent activity and possible
cyberattacks.

FIs can achieve AI supremacy in the financial markets by deploying algorithms that process
alternative data and use the latest machine learning methodologies to outperform their
competitors. FIs should strive for AI transformation for all of their business operations. Some
aspects of the AI revolution in financial services will be focused on reducing costs and
assessing credit default risk more accurately. Compliance functions within FIs can benefit from
AI by providing more efficient ways to comply with the expanding complexity of the regulatory
landscape.

The financial markets are already extraordinarily complex, with hidden feedback loops that are
impossible to sufficiently model. When AI - a complex learning agent - disrupts an already
complicated system, it will impact the overall ecosystem in ways we cannot even begin to
fathom. There will likely be a wave of fortunes made in the financial markets driven by AI. The
quantitative fund industry is only beginning to heat up, and AI traded securities will be the hot
new product on Wall Street and beyond.

One thing is certain - regulators must maintain a watchful eye on the impact of AI on the
financial markets, which will diminish the importance of evaluating the fundamental financial
performance of companies and gradually increase market volatility. Extreme volatility will,
however, accrue quietly in the background, waiting patiently to trigger the next financial crisis.
The role AI may or may not play - whether it will either diminish the impact of that crisis or make
it worse than it might otherwise have been - will remain unknown.

*Daniel Wagner is CEO of Country Risk Solutions. Keith Furst is Managing Director of Data
Derivatives. They are the co-authors of the forthcoming book “AI Supremacy”, which will be
published in September.

This article first appeared in The Asian Banker.