Detecting anomalies with AI: How financial companies can spot risks faster and reduce downtime

SO INTERTWINED and interdependent are today’s complex IT systems spanning the globe that a web server error, a programming glitch, a network issue or, worse, a cyber attack could result in a chain reaction that disrupts not just a website, but also a firm’s core operations.

The stakes are extremely high. These days, a bank service that goes down for hours will draw the ire of customers and cause them to lose their trust. Beyond reputational damage, it could lead to a steep loss of business and financial penalties from government regulators.

In today’s fast-paced and hyper-connected world, businesses and organisations must know how to respond quickly to these issues and prevent them from happening again. To do so, they need to be able to monitor their IT systems in real time. But how can they do that?

The key lies in the need for real-time diagnostics. Consider the warning lights on a car’s dashboard which help pinpoint specific problems in the machine. The driver would be able to get them resolved immediately without opening the bonnet and taking apart the engine to investigate, allowing him to quickly return to the road.

Just as the dashboard provides vital information about the vehicle’s health, companies and organisations need services that offer similar insights into their IT systems. These services should detect anomalies, analyse the root causes and recommend ways to keep operations running without costly downtime.

With the Elastic Search AI Platform, financial companies benefit from the precision of search and the power of artificial intelligence (AI) to find relevant answers fast and in real time.

It allows organisations to proactively resolve observability issues and address complex security threats. IT operations teams can trawl their internal knowledge bases for organisational data, as well as detect, analyse and make use of telemetry from complex IT systems to identify issues and offer solutions to remedy.

More than just a bunch of flashing lights, these alerts also explain clearly what needs to be addressed and in which sequence, providing a step-by-step guide to incident response.

The gap in cyber defences

A study by the Cyber Security Agency of Singapore (CSA) revealed that most companies in Singapore lack adequate cyber security measures. Only a third have fully implemented at least three of the five essential measures the agency recommends.

The lack of preparedness is concerning, given that the cost of detecting and escalating data breaches alone rose from US$1.23 million (S$1.65 million) in 2018 to US$1.58 million in 2023, according to a global survey by IBM.

The risk of threat exposure increases with the growing sophistication of tools used by novice criminals, hackers-for-hire and hacktivists to steal data.

Businesses will need equally sophisticated solutions to proactively monitor their IT systems for potential outages and speed up incident response. But legacy systems coupled with multiple IT monitoring tools often slow down this process.

Explains Ravi Rajendran, area vice-president at Elastic: “I’ve been to organisations where, during an outage, the application team is pointing at a network while the network team is pointing at the application. It can take two to three weeks to get to the bottom of what happened, which could be a simple case of an undersized web server.”

With Elastic, businesses can access a centralised dashboard that makes sense of data from various systems, including applications, infrastructure and security tools.

Its search technology first pulls data from different IT systems. Then, with AI and advanced machine learning, Elastic helps identify various signals that represent anomalies in the systems and presents them on the dashboard.

This holistic approach helps cyber security, site reliability engineers and IT operations managers avoid being swamped by a flood of information and saves them the trouble of having to manually comb through a server’s logs during an incident, so they can quickly find and fix the issue.

“IT personnel can know what’s going on, even the unknowns that may be beneath the surface,” explains Rajendran.

“Financial institutions, like banks, need real-time insights into a system’s behaviour to be able to detect an operational pain point which could be a potential security or financial crime incident,” he adds. “They prioritise swift action when dealing with security vulnerabilities as it’s in their interest to protect customers from successful cyber attacks.

“They can’t afford to spend two hours doing a query and waiting for a response as the attack or threat would have ample time to compromise the system.”

Take SoftBank Payment, an online payment service in Japan, for instance. Its IT team created dashboards to monitor abnormal changes in the number of successful and failed payments, colour-coded in green and red respectively. These dashboards also highlight error codes for invalid credit card numbers, expired cards, or blocked and unusable cards.

If the team observes an increase in failed payments and credit card errors, they are able to identify potential fraudulent transactional activity quickly.

By using the same AI capabilities that cyber criminals are exploiting, financial services firms can simplify threat detection and be even more proactive in strengthening their defences.

Customer expectations from modern banking

For banking and finance firms, customer experience is now as critical as cyber security.

Retail banks alone can grow 3.2 times faster when they regularly optimise customer experience, a Forrester study reported.

As digital services become the norm, customers expect seamless and responsive interactions. This is in addition to a personalised banking experience that anticipates their needs and proactively offers customised solutions to whatever problems they may be facing.

As such, data becomes the fuel that powers business outcomes that can shift the needle.

As data moves across various systems, data centres and cloud platforms, organisations must be able to search through volumes of data generated by their IT systems to obtain relevant and meaningful insights or get a hint of what is wrong in seconds.

Businesses can overcome this with a Search AI platform like Elastic’s, which can search vast amounts of data at speed and scale, and use AI to pinpoint the issue to work on. This helps service desk teams address customer experience incidents faster and more proactively.

“What’s unique about Elastic is that we are a single platform that lets IT teams find what they need quickly, no matter how much data they have,” adds Rajendran.

Modern businesses thrive on responsiveness

In a competitive economy, financial services firms are being judged based on their responsiveness. Those that make use of their data more skilfully gain an edge by providing highly personalised customer service.

The ability to search large volumes of data is crucial to finding the right insights and helping customer support teams make informed decisions.

The retail team at the Royal Bank of Canada, for instance, customised an Elastic-powered dashboard to automatically gather information from multiple sources its salespeople access daily.

A simple search on this dashboard offers them the right details to promptly address a client’s query, so they do not have to follow up later and risk a sales opportunity.

Elastic can go a step further by feeding corporate data into a business’ preferred large language model (LLM) to gain a more contextual understanding of a customer’s query and provide a more accurate and relevant answer.

For example, if the bank wishes to recommend an investment plan to a client, its agent can use AI to sift through both publicly available data and its own internal records to offer a product that best suits the client’s risk appetite, among other considerations.

Such a level of customisation helps banking and finance firms to better meet customers’ needs, build their trust and keep them engaged.

“I can see the rapid adoption of generative AI down the road,” Rajendran says. “What’s important is finding the right data sets, whether they are structured or unstructured, and feeding them back to the AI platform to get the best results for the customer.”

Learn how Elastic can help supercharge your business today.