Introduction
The Global Financial Services Industry is in the midst of massive change across three different dimensions – competition, regulation, and digitization.
- First, across different domains of the industry – such as Capital Markets, Retail Banking, Hedge Funds, Wealth Management, Payment Processing, Financial Exchanges etc – a wave of Fintechs and agile competitors are upending business dynamics.
- Second, Regulators are increasingly focusing on financial risk (credit, market, & operational) & compliance (AML & Fraud Detection).
- Third, a wave of digitization is causing financial services firms to increasingly imitate the FANG (Facebook, Amazon, Netflix, Google) of Silicon Valley. The common thread to these three dimensions is the use of AI and Machine Learning. It is the deployment of these technologies across horizontal functions such as trade execution, backtesting, client servicing, capital optimization, consumer/corporate credit decisions, risk measurement, client onboarding & financial advisory etc that will separate the winners from the losers in the decade to come.
While AI techniques have been around for many years, advances in data storage & processing combined with the increased availability of virtually any kind of data has driven the adoption of use cases spanning a wide spectrum from the mundane (recommendation engines, market basket analysis, customer segmentation) to the exotic (deep learning for data monetization, voice and video recognition). Early adopters will find the increased profitability using Data Monetization by the fusing together of previously unrelated data sources. While it is clear that AI can drive an immense amount of business value for banking, the vast majority of institutions are struggling to deploy the technology at scale.
From a CXO standpoint, the next two years will be the most critical for industry incumbents to begin adopting AI into their product and service strategy with a view to disarming competition and accelerating innovation across their massive customer base.
New Challenges Create New Opportunities – How AI Transforms Financial Services
AI can bring about transformative benefits to financial services industry in the following six ways.
- AI approaches can be applied to a wide & rich variety of business challenges thus enabling an organization to achieve outcomes that were not really possible with descriptive analytics. As covered later in this paper, these use cases range from Trade Strategy development & backtesting in Capital Markets to fraud detection & marketing analytics in Retail Banking.
- When deployed strategically – AI algorithms can scale to enormous volumes of data and help Banks reason over diverse data streams thus increasing automation and reducing manual costs. These techniques can take on problems that can’t be managed manually because of the huge amount of data that must be processed.
- AI approaches can predict the results of complex business scenarios by being able to probabilistically predict different outcomes across thousands of variables by perceiving minute dependencies between them. An example in AML compliance is complex social graph analysis to understand which individuals in a given geography are committing fraud and if there are wider fraudster rings operating in those countries.
- AI approaches are also vastly superior at handling fine-grained data of manifold types than can be handled by the traditional approach or by manual processing. The predictive approach also encourages the integration of previously “dark” data as well as newer external sources of data.
- They can also suggest specific business actions (e.g. based on the above outcomes) by mining data for hitherto unknown patterns. The data science approach constantly keeps learning in order to increase its accuracy of decisions
- Data Monetization – AI algorithms can be used to help interpret the mined data to discover solutions to business challenges and new business opportunities/models.
Some of the current and potential use cases of AI and machine learning include:
– Trade Strategy Development & Execution – In Capital Markets space, Investment Banks, Hedge funds, Asset Managers, Pension Funds & Private Equity, Broker-dealers, and other firms are leveraging AI techniques to analyze vast amounts of diverse data such as Market Data, Financial Intelligence and Non-traditional data in addition to traditional fundamental analysis. This with a view to understanding patterns & sentiment indicators in the data that can indicate portfolio selection, trade strategy development across instruments such as fixed income, equities, and commodities.
– Algorithmic Backtesting – The goal of backtesting is to test investment strategies on vast amounts of historical data to understand performance across a range of alternative scenarios. A variety of techniques ranging from neural networks to traditional time series forecasting are used to analyze vast amounts of data.
– Retail Banking – Customer Service – Retail Banking customer service is one of the most fertile areas of applicability for AI & ML. Applications varying from Credit scoring applicants for various products such as mortgages, auto loans, insights on existing customers from a 360-degree perspective, Chatbots, etc. are being used in forward-looking banks.
– Wealth Management – a Robo-advisor is an algorithm based automated investment advisor that can provide a range of Wealth Management services such as Investment Management, Tax Advisory and Retirement Planning. The Robo-advisor can be optionally augmented & supervised by a human adviser.
– Trade Surveillance – The goal of the Market Abuse Regulation is to ensure that regulatory regime stays in lockstep with the tremendous technological progress around trading platforms especially High Frequency Trading (HFT). For instance, the Market Abuse Directive (MAD) aims to ensure that all EU member states adopt a common taxonomy of definitions for a range of market abuse. Trade Data Repositories are being created at both the supervisory as well as the bank level in order to comply with regulation at both North America and the EU level. AI and ML techniques are being extensively applied to detect a range of rogue trader behaviors such as collusion, insider trading and quote stuffing etc.
– Anti Money Laundering (AML) – AML programs in Capital Markets, Retail Banking and Payments extensively deploy rule-based systems or Transaction Monitoring Systems (TMS) which allow an expert system based approach to set up new rules. These rules span areas like monetary thresholds, specific patterns that connote money laundering & also business scenarios that may violate these patterns. However, fraudster rings now learn (or know) these rules quickly & change their fraudulent methods constantly to avoid detection. Thus there is a significant need to reduce a high degree of dependence on traditional TMS – which are slow to adapt to the dynamic nature of money laundering. AI and ML techniques such as Behavioral modeling & Customer Segmentation can be used to discover transaction behaviors with a view to identifying behavioral patterns of entities & outlier behaviors that connote potential laundering.
– Fraud Detection – The growing popularity of alternative payment modes like Mobile Wallets (e.g Apple Pay, Chase and Android Pay) are driving increased payment volumes across both open loop and closed loop payments. Couple this with in-app payments (e.g Uber) as well as Banking providers Digital Wallets only drives increased mobile payments. Retailers like Walmart, Nordstrom, and Tesco have been offering more convenient in-store payments. This relentless & secular trend towards online payments is being clearly seen in all forms of consumer and merchant payments across the globe. This trend will only continue to accelerate in 2018 as smartphone manufacturers continue to produce devices that have more on-screen real estate. This will drive more mobile commerce. With IoT technology taking center stage, the day is not long off when connected devices (e.g. wearables) make their own payments. However, the convenience of online payments confers anonymity which increases the risk of fraud. Most existing fraud platforms were designed for a previous era – of point of sales payments – with their focus on magnetic stripes, chips and EMV technology. Online payments thus present various challenges that Banks and Merchants did not have to deal with on such a large scale. Fraud detection is now an AI domain and algorithms spanning classical machine learning to neural networks are being leveraged to fight fraud.
– Data Products/Data Monetization – Enterprises operating in the financial services and the insurance industry have typically taken a very traditional view of their businesses. As waves of digitization have begun slowly upending their established business models, firms have begun to recognize the importance of harnessing their substantial data assets which have been built over decades. These assets include fine-grained data about internal operations, customer information and external sources (as depicted in the below illustration). So what does the financial opportunity look like? PwC’s Strategy & estimates that the incremental revenue from monetizing data could potentially be as high as US$ 300 billion every year beginning 2019. This is across all the important segments of financial services – capital markets, commercial banking, consumer finance & banking, and insurance. FinTechs are also looking to muscle into this massive data opportunity. AI approaches are the true differentiators and the key ingredients in any data monetization strategy.
– Cyber Security – Enterprise business is built around data assets and data is the critical prong of any digital initiative. For instance, Digital Banking platforms & Retail applications are evolving to collections of data based ecosystems. These need to natively support loose federations of partner applications, regulatory applications which are API based & Cloud native. These applications are mostly microservice architecture based & need to support mobile clients from the get-go. Owing to their very nature in that they support massive amounts of users & based on their business criticality, these tend to assume a higher priority in the overall security equation. Banks now understand that leveraging real-time analytics based on AI & ML is the foundation of any security strategy. This is only possible by adopting data analytics that provide real-time analysis at extremely low time latencies. This implies an ability to constantly ingest and analyze data from network devices, malware sources, identity and authentication systems. Banks then leverage machine learning and data science to do threat classification as opposed to strict rules-based approaches to analyze relationships between data. Finally, integrating these analytics into these applications such that automatically learning new threat patterns.
Conclusion
The next post in the “AI in Financial Services” will discuss the application of these techniques to Portfolio Backtesting.
1 comment
Good read. thank for sharing.