Large language models (LLMs)-based chatbots are increasingly being adopted in the financial domain, particularly in digital banking, to handle customer inquiries about products such as deposits, savings, and loans. However, these models still exhibit low accuracy in core banking computations-including total payout estimation, comparison of products with varying interest rates, and interest calculation under early repayment conditions. Such tasks require multi-step numerical reasoning and contextual understanding of banking products, yet existing LLMs often make systematic errors-misinterpreting product types, applying conditions incorrectly, or failing basic calculations involving exponents and geometric progressions. However, such errors have rarely been captured by existing benchmarks. Mathematical datasets focus on fundamental math problems, whereas financial benchmarks primarily target financial documents, leaving everyday banking scenarios underexplored. To address this limitation, we propose BankMathBench, a domain-specific dataset that reflects realistic banking tasks. BankMathBench is organized in three levels of difficulty-basic, intermediate, and advanced-corresponding to
Mobile banking apps have transformed the banking sector by offering customers with convenient, secure and easily accessible financial services. Even so, it is crucial for banks and the mobile banking apps developers to understand the factors that influence the utilisation of these apps among Malaysian consumer. This study will examine the influence of several factors which are security concerns, service quality, technological factors and convenience, on the usage of mobile banking apps. The study aims to discover the key factors that affect the usage of mobile banking apps. A quantitative research method was utilised, which involves the collection of data from an online survey. The survey managed to collect data from 152 respondents who are above 18 years old and users of mobile banking apps in Malaysia. The data was analysed with correlation analyses to examine the relationship between the variables. A multinominal logistic regression model was used as a predictive model to predict the usage of mobile banking apps. This study contributes to existing researches by highlighting the importance of security and convenience into the development and marketing strategies of mobile banking
This paper proposes a voice-powered AI-based banking system based on Google Conversational Agent, Dialogflow CX, which provides safe and convenient banking by phone. The system supports essential banking functions such as balance inquiries, transaction history retrieval, card activations, PIN-based authentication of sensitive tasks, smooth live agent handoff for complex and out-of-scope queries, and ensures seamless handover to human agents when required. These tests were performed with high-duration calls, high concurrency, and noisy environments; the system proved to be scalable, responsive, and resilient. All the data used is safely stored in the cloud environment for efficiency and security in real-time voice interactions. A voice-based banking solution that is efficient and easy to use can be provided through this.
Internet banking is changing the banking industry, having the major effects on banking relationships. Banking is now no longer confined to the branches were one has to approach the branch in person, to withdraw cash or deposit a cheque or request a statement of accounts. In true Internet banking, any inquiry or transaction is processed online without any reference to the branch (anywhere banking) at any time. Providing Internet banking is increasingly becoming a "need to have" than a "nice to have" service. The net banking, thus, now is more of a norm rather than an exception in many developed countries due to the fact that it is the cheapest way of providing banking services. This research paper will introduce you to e-banking, giving the meaning, functions, types, advantages and limitations of e-banking. It will also show the impact of e-banking on traditional services and finally the result documentation.
The rapid growth of mobile banking (m-banking), especially after the COVID-19 pandemic, has reshaped the financial sector. This study analyzes consumer reviews of m-banking apps from five major Canadian banks, collected from Google Play and iOS App stores. Sentiment analysis and topic modeling classify reviews as positive, neutral, or negative, highlighting user preferences and areas for improvement. Data pre-processing was performed with NLTK, a Python language processing tool, and topic modeling used Latent Dirichlet Allocation (LDA). Sentiment analysis compared methods, with Long Short-Term Memory (LSTM) achieving 82\% accuracy for iOS reviews and Multinomial Naive Bayes 77\% for Google Play. Positive reviews praised usability, reliability, and features, while negative reviews identified login issues, glitches, and dissatisfaction with updates.This is the first study to analyze both iOS and Google Play m-banking app reviews, offering insights into app strengths and weaknesses. Findings underscore the importance of user-friendly designs, stable updates, and better customer service. Advanced text analytics provide actionable recommendations for improving user satisfaction and expe
This paper investigates the impact of banking competition on interest rates for household consumption loans in the Euro Area from 2014 to 2020. Utilizing a panel data regression approach, we analyze how various factors, including local banking competition, influence the interest rates set by banks across 13 Euro-area countries. Our key independent variable, local banking competition, is measured by the number of commercial bank branches per 100,000 adults. Control variables include the ECB interest rate, euro exchange rate, real GDP growth rate, inflation rate, unemployment rate, bank business volumes, and country risk. We address potential endogeneity and heterogeneity biases and employ both Fixed Effects and Hausman-Taylor models to ensure robust results. Our findings indicate that higher local banking competition is associated with a slight increase in interest rates for household loans. Additionally, factors such as ECB interest rate, country risk, and euro appreciation significantly affect interest rates. The results offer insights into how competitive dynamics in the banking sector influence borrowing costs for households, providing valuable implications for policymakers and
We examine the role of cybercrime legislation around the world in shaping the stability of the banking system. We compile a novel dataset covering the enactment of cybercrime legislation in 132 developed and developing countries to empirically test this research question. We find that the enactment of cybercrime laws enhances the stability of the banking sector. This key finding holds across a comprehensive suite of robustness tests, including alternative measures of bank stability and model specifications. We document significant cross-sectional heterogeneity, with the effect being more pronounced in countries with heavier penalties for illegal cyber activities and legal frameworks that hold banks accountable for their cybersecurity practices. In addition, the positive impact is stronger in jurisdictions with greater international legal cooperation and effective enforcement mechanisms. We further investigate two channels (i.e., funding liquidity and operational risk) through which cybercrime laws may influence bank stability. Our results indicate that these laws can significantly bolster bank stability by enhancing funding liquidity and mitigating operational risk. Overall, our st
This research presented an empirical investigation of the determinants of the net interest margin in Turkish Banking sector with a particular emphasis on the bank ownership structure. This study employed a unique bank-level dataset covering Turkey`s commercial banking sector for the 2001-2012. Our main results are as follows. Operation diversity, credit risk and operating costs are important determinants of margin in Turkey. More efficient banks exhibit lower margin and also price stability contributes to lower margin. The effect of principal determinants such as credit risk, bank size, market concentration and inflation vary across foreign-owned, state-controlled and private banks. At the same time, the impacts of implicit interest payment, operation diversity and operating cost are homogeneous across all banks
Financial regulators such as central banks collect vast amounts of data, but access to the resulting fine-grained banking microdata is severely restricted by banking secrecy laws. Recent developments have resulted in mechanisms that generate faithful synthetic data, but current evaluation frameworks lack a focus on the specific challenges of banking institutions and microdata. We develop a framework that considers the utility and privacy requirements of regulators, and apply this to financial usage indices, term deposit yield curves, and credit card transition matrices. Using the Central Bank of Paraguay's data, we provide the first implementation of synthetic banking microdata using a central bank's collected information, with the resulting synthetic datasets for all three domain applications being publicly available and featuring information not yet released in statistical disclosure. We find that applications less susceptible to post-processing information loss, which are based on frequency tables, are particularly suited for this approach, and that marginal-based inference mechanisms to outperform generative adversarial network models for these applications. Our results demonst
The fourth industrial revolution promotes the integration of Information Technology (IT) and strategic resources. New IT demands and uses have been leading to changes in business processes and corporate governance. Lately, the financial industry has adopted a new integrated banking model known as Open Banking (OB) and the advent of cryptocurrencies has led to the Digital Economy (DE) materialization. Considering these facts, this paper expects to point out through literature review some IT enabling factors that allow the conception of a new industry design (or governance) specifically in the financial industry illustrated by the cases of the Open Banking and Digital Economy. This paper is structured mostly on literature review, accompanied by results, discussions, and finally, conclusions are presented. It was found five potential enabling factors. Keywords: Digital Economy, Information Technology (IT), Open Banking.
We employ a comprehensive data set and a variety of methods to provide evidence on the magnitude of large banks' funding advantage in Canada in addition to the extent to which market discipline exists across different securities issued by the Canadian banks. The banking sector in Canada provides a unique setting in which to examine market discipline along with the prospects of proposed reforms because Canada has no history of government bailouts, and an implicit government guarantee has been in effect consistently since the 1920s. We find that large banks have a funding advantage over small banks after controlling for bank-specific and market risk factors. Large banks on average pay 80 basis points and 70 basis points less, respectively, on their deposits and subordinated debt. Working with hand-collected market data on debt issues by large banks, we also find that market discipline exists for subordinated debt and not for senior debt.
The inception of AI-based fraud detection systems has presented the banking sector across the globe the opportunity to enhance fraud prevention mechanisms. However, the extent of adoption in Nigeria has been slow, fragmented, and inconsistent due to high cost of implementation and lack of technical expertise. This study seeks to investigate extent of adoption and determinants of AI-driven fraud detection systems in Nigerian banks. This study adopted a cross-sectional survey research design. Data were extracted from primary sources through structured questionnaire based on 5-point Likert scale. The population of the study consist of 24 licensed banks in Nigeria. A purposive sampling technique was used to select 5 biggest banks based on market capitalization and customer base. The Ordered Logistic Regression (OLR) model was used to estimate the data. The results showed that top management support, IT infrastructure, regulatory compliance, staff competency and perceived effectiveness accelerate the uptake of AI-driven fraud detection systems adoption. However, high implementation cost discourages it. Therefore, the study recommended that banks should invest in modern and scalable IT s
This paper presents Ryt AI, an LLM-native agentic framework that powers Ryt Bank to enable customers to execute core financial transactions through natural language conversation. This represents the first global regulator-approved deployment worldwide where conversational AI functions as the primary banking interface, in contrast to prior assistants that have been limited to advisory or support roles. Built entirely in-house, Ryt AI is powered by ILMU, a closed-source LLM developed internally, and replaces rigid multi-screen workflows with a single dialogue orchestrated by four LLM-powered agents (Guardrails, Intent, Payment, and FAQ). Each agent attaches a task-specific LoRA adapter to ILMU, which is hosted within the bank's infrastructure to ensure consistent behavior with minimal overhead. Deterministic guardrails, human-in-the-loop confirmation, and a stateless audit architecture provide defense-in-depth for security and compliance. The result is Banking Done Right: demonstrating that regulator-approved natural-language interfaces can reliably support core financial operations under strict governance.
Banking Transaction Flow (BTF) is a sequential data found in a number of banking activities such as marketing, credit risk or banking fraud. It is a multimodal data composed of three modalities: a date, a numerical value and a wording. We propose in this work an application of self-attention mechanism to the processing of BTFs. We trained two general models on a large amount of BTFs in a self-supervised way: one RNN-based model and one Transformer-based model. We proposed a specific tokenization in order to be able to process BTFs. The performance of these two models was evaluated on two banking downstream tasks: a transaction categorization task and a credit risk task. The results show that fine-tuning these two pre-trained models allowed to perform better than the state-of-the-art approaches for both tasks.
Propelled by the recent financial product innovations involving derivatives, securitization and mortgages, commercial banks are becoming more complex, branching out into many "nontraditional" banking operations beyond issuance of loans. This broadening of operational scope in a pursuit of revenue diversification may be beneficial if banks exhibit scope economies. The existing (two-decade-old) empirical evidence lends no support for such product-scope-driven cost economies in banking, but it is greatly outdated and, surprisingly, there has been little (if any) research on this subject despite the drastic transformations that the U.S. banking industry has undergone over the past two decades in the wake of technological advancements and regulatory changes. Commercial banks have significantly shifted towards nontraditional operations, making the portfolio of products offered by present-day banks very different from that two decades ago. In this paper, we provide new and more robust evidence about scope economies in U.S. commercial banking. We improve upon the prior literature not only by analyzing the most recent data and accounting for bank's nontraditional off-balance sheet operation
This paper discusses the security posture of Android m-banking applications in Qatar. Since technology has developed over the years and more security methods are provided, banking is now heavily reliant on mobile applications for prompt service delivery to clients, thus enabling a seamless and remote transaction. However, such mobile banking applications have access to sensitive data for each bank customer which presents a potential attack vector for clients, and the banks. The banks, therefore, have the responsibility to protect the information of the client by providing a high-security layer to their mobile application. This research discusses m-banking applications for Android OS, its security, vulnerability, threats, and solutions. Two m-banking applications were analyzed and benchmarked against standardized best practices, using the combination of two mobile testing frameworks. The security weaknesses observed during the experimental evaluation suggest the need for a more robust security evaluation of a mobile banking application in the state of Qatar. Such an approach would further ensure the confidence of the end-users. Consequently, understanding the security posture would
This paper develops a dynamic monetary model to study the (in)stability of the fractional reserve banking system. The model shows that the fractional reserve banking system can endanger stability in that equilibrium is more prone to exhibit endogenous cyclic, chaotic, and stochastic dynamics under lower reserve requirements, although it can increase consumption in the steady-state. Introducing endogenous unsecured credit to the baseline model does not change the main results. The calibrated exercise suggests that this channel could be another source of economic fluctuations. This paper also provides empirical evidence that is consistent with the prediction of the model.
With the growing competition in banking industry, banks are required to follow customer retention strategies while they are trying to increase their market share by acquiring new customers. This study compares the performance of six supervised classification techniques to suggest an efficient model to predict customer churn in banking industry, given 10 demographic and personal attributes from 10000 customers of European banks. The effect of feature selection, class imbalance, and outliers will be discussed for ANN and random forest as the two competing models. As shown, unlike random forest, ANN does not reveal any serious concern regarding overfitting and is also robust to noise. Therefore, ANN structure with five nodes in a single hidden layer is recognized as the best performing classifier.
Today, almost all banks have adopted ICT as a means of enhancing their banking service quality. These banks provide ICT based electronic service which is also called electronic banking, internet banking or online banking etc to their customers. Despite the increasing adoption of electronic banking and it relevance towards end users satisfaction, few investigations has been conducted on factors that enhanced end users satisfaction perception. In this research, an empirical analysis has been conducted on factors that influence electronic banking user's satisfaction perception and the relationship between these factors and the customer's satisfaction. The study will help bank industries in improving the level of their customer's satisfaction and increase the bond between a bank and its customer.
We study the causes and consequences of bank runs. By applying large language models to historical newspapers, we create a comprehensive database of bank runs in U.S. history with information on 3,984 runs on individual banks from 1863 to 1934. Our novel data allow us to establish that runs are considerably more likely in weak banks but also occur in strong banks, especially in response to negative news about the real economy or the broader banking system. However, runs typically only result in failure for banks with poor fundamentals. Strong banks survive runs through various mechanisms, including signaling strength, interbank cooperation, and temporary suspension. At the local level, runs on banks with poor fundamentals translate into substantially larger declines in deposits, lending, and manufacturing activity than runs on strong banks. Our findings imply that poor fundamentals are central to explaining both when runs occur and when they have severe economic effects, tempering the view that small shocks can generate discontinuous jumps to bad equilibria through self-fulfilling run dynamics.