A couple of years after its initial boom, artificial intelligence (AI) still remains a huge buzzword in the fintech industry, as every firm looks at a new way of integrating the tech into its infrastructure to gain a competitive edge. Exploring how they are going about doing this in 2025, The Fintech Times is spotlighting some of the biggest themes in AI this February.
Having explored the impact of AI on customer service and how machine learning can both help and hinder financial decision-making, we now turn our attention to back-office operations and how AI can be properly utilised. If it is seen as the solution to all problems, a company is bound to fail: exploring the potential drawbacks of automating back-office operations, we reached out to the industry.
Overextension causes problems
Exploring this very issue of addressing every issue with AI, Brendan Thorpe, customer success manager at Auriga, the omnibanking platform provider, says: “The key driver for a lot of banks to automate the back office is to improve on the legacy systems and processes that often drive them. With that aim is often a desire to reduce costs and within that headcount, however, the people that manage those systems and processes often form an integral part of the system and the checks and balances that exist within it.
“Many of the key drawbacks lie in overextending automation such that you lose the capabilities that are required to maintain those checks and balances many of which require manual intervention.
“Another critical area to consider lies in the need to collate the data that enables you to automate those processes. Older systems are often separated into siloes, which can make that a complex process and often means pulling it into a data lake. This means that as banks automate back-office operations, they also increase their attack surfaces and create vulnerabilities that can be exploited.
“To address this, financial services organisations must implement zero trust strategies, ensuring that nothing is executed unless it has been previously authorised. They must establish robust network security policies to build resiliency and protect themselves. Layered protection and continuous monitoring for attacks is important to ensure that the organisation is fully protected and remains resilient.”
Is the cost worth it?
From a financial standpoint, automation can also be extremely draining on resources. Hamish Warnock, senior relationship manager at Canoe Intelligence, the alternative investment experts, explores how, when improperly used, AI can be a drain on a company’s resources as they need to commit funds and minds to fix the problem created by poor automation.
“Humans are creatures of habit. Teams will often be reticent to implement change, particularly when it involves introducing new technology and entirely overhauling familiar processes. For some, this may be daunting, as it requires learning new systems and workflows. Teams may also be concerned about job displacement as automation may reduce, or eliminate, the need for certain administrative roles.
“On the flip side, by automating processes, firms free up staff from manual time-consuming, repetitive tasks, and can reallocate them to more valuable and engaging work which contributes more directly to the firm’s success.
“Teams may be concerned that automation will lead to less human interactions, potentially weakening company culture. This fear may be unfounded. Back-office activities often involve nuances unique to each client, requiring human judgement which technology can’t yet replicate. The human element remains, for now, vital to the smooth running of back-office activities.
“There is a cost associated with automation, so the investment extends beyond just the technology to include third-party support to integrate the system and train staff.
“However, this investment can translate into savings over the long run as automation enhances efficiencies and reduces operational costs – freeing up capital for innovation and company expansion.”
Keeping humans in the loop
Focusing on the human touch employees bring, John O’Connell, founder and CEO at The Oasis Group, the growth consultancy, notes that automation cannot work to its best level without oversight.
“One of the highest drawbacks is the risk of trusting AI too much. Back-office operations team members need to maintain a human in the loop to ensure that AI processing of information is accurate.
“For example, consider a client service associate in a wealth management firm who needs to enter information from a potential client’s existing investment holdings to develop a financial plan. AI can be used to interrogate the PDF file from the prospect’s existing investment portfolio. This can save hours of data entry for the CSA.
“However, the CSA should check the data before entering it into their financial planning software. A small error in AI’s processing of the PDF may transpose a letter in the ticker symbol of one of the holdings. This can lead to an inaccurate financial plan.”
Automation can succeed or fail based on the employees
Chris McGee, managing director at AArete, the global management and technology consulting firm, shared similar views, stating that depending on how employees are used following the introduction of automation, they can either be very helpful to an organisation or be a major issue for potential losses.
“One of the biggest potential drawbacks of automating back-office operations is the people side of the equation. Automation opportunities continue to grow, which in turn should benefit roles for employees across the middle and back-offices.
“Oftentimes, this enables them to shift focus from manual tasks to more strategic priorities, which is clearly a benefit. However, in instances where automation is introduced and clear change management processes have not been undertaken, certain employees can get lost in the shuffle.
“If they are not reassigned to more strategic initiatives or not provided with proper training on new roles, they risk losing their positions. Similarly, employees can become too dependent on automation, so much so that when automations fails or issues occur in the process, employees may not be able to manually take over these processes given they have become too unfamiliar with the automated process and its intended steps in the first place!”
Loss of control and accountability
Dave Csiki, managing director at INDATA, the investment management software solution provider, looks at the important role automation can play in back-office operations. However, he notes that without the right control, these automated processes won’t work as they should. He said: “The potential drawbacks of automating back-office operations are the potential loss of control and accountability over important tasks like data reconciliation which goes into accurate and timely reporting, which is the key output of back-office operations.
“Automation tools and partners need to be able to provide effective, yet flexible SLAs (service level agreements) that specify and enforce the rules of engagement for all key processes that are used to automate back-office operations.”
Echoing this point, David Wyke, head of financial solutions at Codec UK, the IT solutions consultancy, also noted the key role context has in impacting the success of automation.
“Historically, machines have missed nuances and context, which has led to responses feeling rigid or
impersonal. As AI has evolved, sentiment analysis has greatly improved, allowing AI to provide
responses with the correct tonal feel. There is a clear benefit in using AI to automate repetitive low-
value-added tasks, as this allows individuals in organisations to work on high-value-added activities
and creates bandwidth to focus on more strategic and operational goals.”
The post What Are the Potential Drawbacks of Automating Back-Office Operations? appeared first on The Fintech Times.
