In corporate banking, risk management strives to limit the risk exposure and asset losses for a financial institution. It can be extremely complicated, and it requires sophisticated data analytics that is increasingly real time. Its scope is very wide, and it extends throughout all of the bank’s different businesses. Key risk management areas of interest include (and this is not exhaustive) fraud, investment, trading, margin and derivatives exposure, payment risk, credit exposure, debt levels and liquidity to meet day-to-day and ongoing obligations, regulatory compliance, and financial market exposure (e.g., investments, foreign exchange exposure).
When risk management falls short, it can lead to billions of dollars in losses and reputational damage. As risk can happen across many departments, it’s difficult for auditors and risk managers to catch problems early without proper controls and stress testing.
For example, a federal judge last year ruled that Citigroup is not entitled to recoup $893 million it accidentally wired to Revlon, saying it was “a banking error of perhaps unprecedented nature and magnitude.” It was another blow to Citigroup, which received a $400 million fine in 2020 for “longstanding failure to establish effective risk management.”
In another well-known example, the failure of Archegos Capital Management last year led to more than $10 billion in losses, including $5.5 billion in losses for Credit Suisse and a nearly $3 billion loss for Japanese bank Nomura Holdings. Last December, the Federal Reserve Board provided additional guidance to banks of its expectations regarding risk management practices in investment banking.
These types of financial losses highlight the need for improved corporate bank risk management, especially in the face of increasing competitive pressures and regulatory oversight.
Using AI to Extract Valuable Insights in Risk Management
To manage risks in real time and make intelligent decisions, financial institutions over the next decade will continue to prioritize advanced analytics by using artificial intelligence (AI) systems to extract deeper insights. The most advanced banks are starting to utilize neural nets and deep learning, which can ingest millions of data points in milliseconds to detect problems. According to McKinsey’s research, the percentage of a corporate bank’s risk management staff focused on analytics will increase from 15% to 40% by 2025.
Corporate banks can use AI to determine high-risk areas and provide automation and controls to limit the risk. AI can identify patterns and predict outcomes to help banks understand and mitigate risk more effectively. AI can help corporate banks strategize for the future, make precise real-time decisions, improve risk modeling, provide better monitoring, and minimize costly human errors.
To accomplish this, there are three key requirements AI systems need for data scientists to select, tune, and build the best algorithms. First, they need to use massive volumes of data to learn and then improve and optimize information for an organization. Second, AI systems need to consume multiple data sources, such as transactional, account, customer, payments, and various third-party data, often at the edge or from different data silos or geographies. Third, AI systems need a hyper-capable database that can ingest and process all this data fast, as in milliseconds, to make decisions in real time.
Many banks still use traditional data platforms with inconsistent and incomplete datasets from disparate sources that are hard to extract and act in batch mode. For banks that require a more capable, real-time approach, a modern database engine is needed.
For example, a leading multinational financial services company moved to a modern data platform to accurately manage in real time account authentication, trade authorization, and compliance/risk controls. The data platform handles large amounts of data quickly, ensuring that the company provides best-in-class responsiveness to customers’ trading activities while remaining in compliance with securities regulations and internal controls. At the same time, it ensures consistent data and performance with scalability and low latency, even during peak trading periods.
Financial institutions are susceptible to risk due to the sensitive information they collect. Advanced analytics and automation are reshaping the way risk is managed, and it’s no surprise that the leading firms are moving to sophisticated AI-based solutions. With more corporate banks facing unprecedented worldwide regulatory and market pressures, relying on AI will help automate processes to minimize costly human errors and provide greater visibility and insight into the critical risk categories. To meet these goals, a modern, real-time data platform that can ingest, process, and deliver sophisticated data analytics quickly, reliably, and consistently is critical.
