The banking sector is characterized by a significant presence of legacy technology. This is expensive: nearly two-thirds (64%) of retail banks' IT budget is spent on system maintenance. Of course, this isn’t a trivial matter in banking; these systems hold personal information about millions of customers. With regulation and reputation being significant concerns for large banks, modernizing legacy systems can feel fraught with risk and, perhaps, ultimately unnecessary.
This is why core systems and applications have lacked investment and today are ill-equipped for the demands — and opportunities — of the modern marketplace. However, generative AI (GenAI) can play an important role in providing a robust foundation for banks’ legacy modernization initiatives. Technology's capability to reframe and translate vast amounts of information into new, easily consumable formats is crucial in helping banks initiate and accelerate legacy modernization initiatives.
Given the costs of failing to modernize could be substantial — whether in the form of future maintenance costs or missed opportunities for innovation — GenAI has the capacity to be an integral part of the banking sector’s future. Without it, it may be hard to deliver a legacy modernization initiative: with it, you could unlock commercial advantages fast.
In this article we’ll take a closer look at how banks could make use of GenAI for legacy modernization.
Why do banks struggle with legacy modernization?
Organizations across the board find legacy modernization challenging. However, for banks, there are a number of things that make it uniquely difficult:
High levels of regulation: This means making changes to complex systems can take time and require layers of oversight.
Substantial technical debt: Because of the challenges of updating complex systems, banks are more likely than organizations in other industries to suffer from technical debt.
Security risks: Old systems and high levels of regulation — not to mention the financial power of banking institutions — make tackling security issues both more urgent but also more difficult.
Capability gaps: Very old legacy systems, written in languages like COBOL (common business oriented language), are not well-known today — pair that with domain-specificity, finding people with the right skills and knowledge can be difficult for banks.
While it may be tempting for leaders in these institutions to carry on as normal — if it’s worked all this time, why change things? — this is a mistake. Upstart FinTech companies are changing customer expectations about what financial and banking services and products should be like in a fully digital world, while a changing regulatory environment means you can’t rely on decades of successful compliance — you need to be responsive enough to meet the demands of customers and regulators as they evolve.
This means addressing technical debt quickly, and setting yourself up so you can evolve and adapt is essential.
How banks can leveraging generative AI to accelerate legacy modernization
This is where GenAI comes in. It can’t get rid of the modernization process, but it can accelerate or simplify a number of important steps in the process. As a result, what initially may have looked like a huge expensive obstacle can become a more manageable project. Ultimately, that gives banks an opportunity to unlock the benefits of modernization faster and more efficiently.
Let’s take a look at some of those areas.
1. Accelerating code analysis and reverse engineering
Banks have long wanted to re-engineer or replace old code; converting COBOL to a modern language like Java, for example, is a common goal. However, this is a mammoth task that will require hundreds of engineers and will likely take several years. These legacy systems aren’t trivial — they often power critical functions like transaction processing, customer account management and risk assessment.
To compound the problem, most source code programmers have long since retired, which means it’s difficult (and expensive) to find and recruit them in sufficient numbers. This only adds to the risk of the entire exercise.
Subsequently, the project remains on the back burner. Even if it’s said to be a priority, it’s often hard to justify in an organization’s current circumstances.
Fortunately, GenAI can help banks gain visibility into their existing legacy systems. It can be particularly effective in dissecting and analyzing complex, outdated codebases that are still widely used in banking, such as those written in COBOL, Fortran, or RPG (Report Program Generator) on AS/400 systems. By using GenAI alongside techniques like natural language processing (NLP), it’s possible to quickly create detailed documentation and code summaries of old, opaque systems.
Benefits:
Reduced time and effort: Accelerates the reverse engineering process, reducing time spent on understanding legacy banking systems (e.g core banking platforms) logics which are written decades back and have interdependencies. Cuts the time required for code analysis and reverse engineering by automating these processes, accelerating project timelines.
Increased developer productivity: Allows developers to focus on higher-value tasks rather than spending excessive time deciphering legacy code, enhancing efficiency. This will reduce the dependency on legacy technology developers - old Banking platform developers.
Enhanced documentation: Produces accurate and comprehensive documentation that aids in ongoing maintenance and future modernization efforts, ensuring continuity and compliance.
Knowledge transfer: Bridges knowledge gaps caused by personnel turnover by providing comprehensive system documentation, crucial in the highly regulated banking industry.
2. Improving architectural decision making
Having composable, interoperable and coreless architectures is key to most banks’ modernization, because it allows parts of their legacy applications to co-exist with modernized parts and even third-party products and platforms.
However, many core banking systems are based on monoliths; they’re prone to outages and ill-suited to rapid recovery. That makes life much harder for engineering teams and hampers their ability to develop and deliver new features that would drive value.
By using GenAI to gain visibility into these legacy systems, engineers and architects can tackle architecture problems faster and be more deliberate in the types of solutions they move towards. Ultimately it should allow them to ensure modernization work is properly aligned with the requirements of the bank.
Benefits:
Enhanced decision making: Provides data-driven insights and simulations, leading to better-informed architectural decisions tailored to banking operations.
Time savings: Speeds up the design phase, allowing teams to reach optimal solutions more quickly, minimizing downtime.
Innovative solutions: Enables the exploration of innovative design options that might not have been considered otherwise, such as new financial products or services.
3. Managing risk and improving resilience
Banking legacy systems were not designed to handle the fast moving complex regulatory requirements of today’s financial landscape, such as those related to anti-money laundering (AML), Know Your Customer (KYC), and data protection regulations like GDPR.
Generative AI mitigates risks associated with legacy modernization by enabling comprehensive and rigorous testing programs. It automatically generates extensive test cases and improves system monitoring and maintenance, which is critical for banking systems that handle sensitive financial data.
Benefits:
Reduced vulnerability: Identifies and mitigates security vulnerabilities in legacy systems,enhancing overall security and compliance with financial regulations.
- Thorough testing: Ensures new systems are rigorously tested, reducing the likelihood of bugs and issues post-implementation that could affect banking operations.
- Proactive maintenance: Provides real-time monitoring and maintenance insights, allowing teams to address issues proactively, ensuring system reliability.
Supporting compliance and security
Generative AI helps ensure that new systems comply with regulatory requirements by continuously monitoring and updating compliance protocols. It can also identify and mitigate security vulnerabilities in legacy systems, which is critical for banks to maintain trust and compliance.
Benefits:
Regulatory adherence: Automatically generates compliance-related documentation and reports, ensuring adherence to financial regulations like GDPR, PCI DSS, and others.
Improved security: Identifies potential security risks in legacy systems and suggests remediation steps to secure modernized systems, protecting sensitive financial data.
Continuous compliance monitoring: Provides ongoing monitoring and updates to maintain compliance as regulations evolve, reducing the risk of regulatory breaches.
Generative AI addresses key challenges in legacy modernization for banks, making the process more efficient, secure, and aligned with regulatory requirements while enabling banks to leverage innovative solutions and enhance overall operational effectiveness.
A robust foundation for your legacy modernization initiative
A legacy modernization project is always daunting, but generative AI can help reduce the scale of the task in front of your teams so it can be more easily managed and delivered faster and with greater confidence.
The underlying technology of GenAI helps us to make sense of large volumes of data and information, see connections between different parts of a system and free ourselves from needing to understand an entire codebase in order to start improving it.
In basic terms, this approach works something like this:
Abstract syntax trees break legacy code into logical, readable chunks.
Graphs make it possible to build connections across these readable chunks of code (which helps us query it if we want to find something).
Generative AI (more specifically large language models) to summarize the documentation to non-experts.
Generative AI has captured headlines but it’s time to move beyond the hype. This technology offers powerful, strategic applications that can unlock real value for your organization. Don't miss out — now's the time to explore genAI's potential for serious business impact.