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.
While there are drawbacks to automating back-office operations if the technology is used properly, there are also countless benefits as well. In this article, we turn our attention to the positive side of automating back-office operations and how AI has impacted traditional processes.
Helping decisioning
Charles Clarke, group vice president, Guidewire, the insurance analytics firm, notes that the emergence of AI is massively impacting insurance back-office operations. However, AI is not a solution that can remove humans from the decision-making process. Exploring the tech’s impact on traditional approaches, Clarke said: “Process AI and machine learning have increased their role in how insurance back-office functions like claims processing and front-office functions like underwriting are conducted.
“In the last two years, generative AI has been stepping into the back office, helping to reason and evaluate a claim or cost a risk; its impact in the front office is less well felt.
“Insurers have, however, realised that their business model must remain human-focused and that customers want to deal with a human being when it matters most. The opportunity with claims handling vis-a-vis AI, is not to automate and remove humans, but to automate and make space for human empathy. If a claims handler has more time—courtesy of AI—to engage with their customers, customer satisfaction will increase.
“Our annual evaluation of customer attitudes to how insurers innovate with Insurtech found that over half (53 per cent) of UK customers were uncomfortable with the idea of AI making decisions about the value of their claim without human intervention. So, the evolutionary path for AI proliferation in all parts of an insurance business means playing a greater role in decisioning must go together with defining a clear role for human intervention and guidance.”
Ensuring growth doesn’t mean fraud slips through the cracks
An organisation’s growth can mean more customer onboarding and customer interaction. However, if the correct measures are not put in place, traditional screening might be overwhelmed by potential fraud cases, and in turn let some fall through the gaps. According to Andrew Steadman, chief product officer at SBS, the firm helping the financial sector digitise, this is where AI can help.
“AI is fundamentally transforming back office operations that have traditionally been defined by some of the most heavily manual and time-consuming processes in the finance industry and while process automation has increased productivity to a degree embedding AI can make a significant impact.
“Within fraud detection, for example, AI is giving banks and financial institutions the ability to analyse large swaths of data in seconds to detect unusual patterns and anomalies in real-time.
“By using AI rather than simple rules banks can do a far better job of identifying false positives and identifying those transactions which require some form of human intervention. It allows growth in volumes without a similar growth in investigator capacity while effectively managing risk for the financial institution.
“In regulatory reporting and compliance checks, generative AI tools are also providing financial organisations with major efficiency gains by automating the process of drafting preliminary responses to reporting and compliance requests based on existing organisational data. This automation not only eliminates the time that employees would otherwise spend pulling this data manually, but is also much less prone to human error.”
Disputes and fraud recovery
As a consumer, there is nothing more infuriating than waiting ages for a response from a bank about a fraud case or dispute, only to eventually hear that the claim was not successful. Thomas Müller, co-founder and CEO of paytech Rivero, notes that with AI, these potential problems can be fast-tracked, meaning banks can reallocate resources elsewhere, and consumers get a faster response.
“Through AI and automation, the traditionally slow, manual processes used to power back-office operations have turned into fast and efficient systems. Using banks as an example, back-office automation has been used to reduce time spent on payment disputes, fraud recovery and payment network compliance, leading to a major boost in efficiency.
“Automation in the back-office instead of the customer-facing process has been a real game-changer. Often, a customer-facing digital transformation eventually reveals itself as a “digital facade.” It appears modern and more efficient, but in reality, it simply hides the legacy operation that is still used behind the scenes. Simplifying payment operations in the back office makes banking smoother for the end customer and results in a better user experience.”
Freeing up resources
Traditionally, back-office operations included a lot of repetitive and manual data entry tasks. Stephen Greer, a banking industry consultant at data and AI provider SAS, notes that using AI effectively means employees do not need to be tied down to these tasks and can be used elsewhere.
“Banking is a step ahead of other industries in AI adoption. SAS and Coleman
Parkes released a study last fall that showed banks are using GenAI at a higher rate than other industries in several back-office functions, including IT and finance.
“As far as how AI has changed those operations, I think there are two ways to look at it: process-specific and operational.
“From the process side, we’ve seen a lot of robotic process automation implementation over the last 10 years. The back office traditionally included a lot of repetitive and manual data entry or data replication tasks that have been automated by RPA.
“You have a lot better machine learning and AI algorithms now to look at back-office data to find issues, anomalies, fraud and financial crime.
“In the future, generative AI will be used to automate document processing and handling of unstructured data – even payments data, where text-based flat files are still used.
“On the operational side, there has been a lot of improvement on the monitoring and delivery of back-office functions. Automated quality checks, error detection and processing time.
“Over time, with generative AI, I think you’ll start to see more back-office personnel implementing things like Copilot to help provide leverage.”
Removing time-intensive, repetitive tasks off the to-do list
Sharing similar views, Nik Talreja, CEO of Sydecar, a deal execution platform for venture investors, added: “AI is in its early innings, and its impact on back-office operations is starting to unfold. Several new products focused on back-office operations are well-poised to solve repetitive, time-intensive tasks that require little more than process adherence.
“A few examples: analysing negotiated legal forms to understand proposed edits and draft counter-offers in line with company policies, reviewing bulk financial information for discrepancies, forecasting cash flows using a broad basket of unstructured data, and cleaning up databases based on natural language processing (NLP).
“NLP has the potential to extract terms from subscription agreements and summarise investment updates to auto-generate investor statements. While use cases are endless, current adoption is limited to applications that can be easily integrated in traditional workflows. For this reason, I believe that incumbent service (software) providers are best poised to win, as they own distribution.”
Data analysis
Radhakrishnan Srikanth, SVP of engineering and AI at Tyfone, the digital banking and payments solutions provider, also shared his enthusiasm at AI’s ability to alleviate resources. However, he also added that another application of the technology lies in data analysis
“Financial institutions (FIs) are inundated with customer data, but it’s time-consuming to process the information manually. AI detects potential fraud in real-time, flagging potential anomalous activity as it occurs. AI-driven data analysis also empowers FIs to better market to your consumers. In the time it takes to analyse the data, your customers’ needs may have changed.
“With AI and natural language understanding, you can automatically summarise, categorise and visualise the data to provide better insights into what your customers are doing and what they need from your FI. This enhances visibility into account holder behaviour and trends, enabling FIs to better serve their customers by anticipating their needs.
“Digital banking platforms are now incorporating more AI-driven tools to improve user experiences. For instance, Tyfone’s nFinia digital banking platform is powered by Penni AI, an AI-powered financial concierge that provides real-time support, streamlining account holder interactions and easing the burden on staff by lowering support calls and queries.
“Powered by natural language, Penni AI seamlessly guides users through their banking experiences which makes a huge difference on the back-office. This is just the beginning of how AI is transforming fintech, and I anticipate these innovations to continue to accelerate.”
