Artificial intelligence in financial services: enhancing efficiency, risk management, and customer experience

Liu Mingzhu1
1 Patrice Lumumba Peoples\' Friendship University of Russia

Статья в журнале

Вопросы инновационной экономики (РИНЦ, ВАК)
опубликовать статью | оформить подписку

Том 15, Номер 3 (Июль-сентябрь 2025)

Цитировать эту статью:

Аннотация:
In the context of economic digitalization, the application of artificial intelligence (AI) in the financial sector is becoming an essential element of its transformation. This article explores key areas of AI implementation in financial services, including risk management, customer service personalization, and the optimization of business processes. Special attention is given to the comparative analysis of regulatory and technological approaches in the United States, China, and the European Union. The scientific novelty of the study lies in the systematization of risks and opportunities associated with AI use in finance, as well as in the development of recommendations to ensure transparency and resilience of AI-based solutions. The research is based on secondary data analysis, legal document review, and analytical reports. The findings will be of interest to researchers in financial technology, regulators, AI developers, and financial sector practitioners seeking to safely and effectively integrate AI into their professional activities.

Ключевые слова: AI, fintech, risk, management, customer relations, financial services

JEL-классификация: O31, O32, O33, E44, G00



Introduction

Amid the rapid advancement of global digital transformation, artificial intelligence (AI) has emerged as a pivotal driver of high-quality economic progress. This impact is especially pronounced in the financial services industry, where AI demonstrates significant transformative potential. Given the sector’s inherent reliance on data analysis, information flow, and risk management, it is naturally well-positioned to benefit from AI integration. Technologies such as big data analytics, machine learning, and natural language processing are reshaping financial institutions’ operations, organizational models, and modes of customer engagement.

AI is playing an increasingly vital role in enhancing operational efficiency, optimizing user experiences, improving regulatory compliance, and minimizing human-related errors. For instance, in the banking sector, AI-powered credit scoring systems are adept at identifying high-risk borrowers. Automated investment platforms (robo-advisors) deliver customized portfolio management services, while advanced anti-money laundering tools leverage deep learning to detect suspicious activities and prevent financial misconduct.

On a broader scale, leading global economies recognize AI’s strategic importance in reshaping the financial landscape. The United States, through market-driven initiatives and its National AI Strategy, supports the rapid evolution of fintech enterprises. China, following the launch of its "New Generation Artificial Intelligence Development Plan" in 2017, has prioritized financial intelligence within its broader technology roadmap, fostering collaboration across government, industry, academia, and application sectors. Meanwhile, the European Union champions data ethics and protection through its "trustworthy AI" framework and continues to advance the legislative development of the AI Act.

Nevertheless, despite AI’s growing relevance in finance, several challenges persist. Issues such as lack of algorithmic transparency, system robustness, data security, and cross-border regulatory alignment remain unresolved. Researchers have highlighted that AI models often lack resilience under extreme market conditions and may exhibit biases when relying on historical datasets. Additionally, real-world deployment of these technologies can introduce new concerns, including discrimination, systemic risk, and outdated regulatory responses.

In conclusion, analyzing the application, risks, and governance of AI in the financial domain is both theoretically important and practically urgent. As financial technologies continue to evolve, understanding how innovation and institutional design can harmoniously advance will be crucial to building a smarter, safer, and more resilient financial ecosystem.

In recent years, the application of artificial intelligence in financial services has attracted increasing scholarly and industry attention, leading to multifaceted discussions on how AI enhances operational efficiency, strengthens risk management, and optimizes client engagement.

According to Zhao et al. (2023), incorporating machine learning into commercial banks’ credit approval procedures significantly boosts the precision of credit evaluations and helps lower the rate of non-performing loans [1, p. 24–36]. Similarly, Chen & Liu (2023) explored AI's influence on portfolio strategies, noting that AI systems demonstrate superior responsiveness and data processing capabilities in large-scale financial environments, particularly in high-frequency trading contexts [2, p. 78–89].

From the standpoint of risk mitigation, Wang et al. (2024) emphasized the increasing relevance of AI-powered anti-fraud technologies for real-time monitoring of suspicious activities, especially in the expanding realm of mobile payments [3, p. 55–70]. Smith & Taylor (2023) conducted a survey across European financial institutions and found that AI substantially improved the robustness of compliance systems and internal anti-money laundering (AML) frameworks [4, p. 103–118].

Concerns regarding ethics and data management are also gaining prominence in the literature. Lee & Kim (2023) investigated algorithmic bias, warning that AI-driven credit scoring may result in discriminatory outcomes due to imbalanced datasets, and they advocated for transparent and equitable financial AI governance [5, p. 15–32]. From a regulatory angle, Nguyen et al. (2024) analyzed the capability of AI systems to comply with data protection laws, particularly under the EU’s General Data Protection Regulation (GDPR), and stressed the importance of aligning privacy safeguards with technological innovation [6, p. 41–56].

Cross-national comparisons further enrich the discourse. Huang et al. (2024) assessed the approaches taken by China, the U.S., and the EU, highlighting that China favors a "policy-led and tech-integrated" model, the U.S. leans towards "innovation-first with market mechanisms", and the EU prioritizes "ethical oversight and risk mitigation" [7, p. 60–74].

On the technological application front, Garcia et al. (2023) introduced a chatbot system based on natural language processing (NLP), which demonstrates effective usage in sectors such as customer support, insurance processing, and investment consulting [8, p. 88–99]. Kumar & Singh (2023) focused on the potential of AI for credit scoring in microfinance, asserting that it significantly enhances the efficiency and inclusivity of lending services [9, p. 22–38].

Nevertheless, the limitations of AI also warrant attention. Jin & Zhang (2023) warned that AI models heavily trained on historical data may underperform during unpredictable “black swan” events, urging financial institutions to develop robust multi-layered risk defense mechanisms [10, p. 127–140]. Xu et al. (2024), through empirical research, validated AI’s positive effect on risk detection in small and medium-sized banks, underscoring that the accuracy of outcomes is largely dependent on data quality and the interpretability of models [11, p. 99–115].

In addition, Park and Lee (2023) examined the dynamics of human-computer interaction in AI-powered financial customer service. Their study highlights the considerable advantages of AI-based chatbots in enhancing response speed and user satisfaction within banking services, particularly in the retail sector. These systems also demonstrate notable scalability and cost-efficiency benefits [12, p. 51–66].

Torres and Salazar (2023) investigated the integration of artificial intelligence and blockchain technologies within the framework of regulatory technology (RegTech). They argue that this convergence can significantly lower compliance costs and improve regulatory transparency, particularly in the monitoring of cross-border financial transactions [13, p. 118–132].

Zhao and Wen (2024) focused on the role of AI in optimizing financial decision-making processes. They propose that AI models incorporating multi-objective optimization algorithms can enhance the efficiency of asset allocation, offering financial institutions a novel approach to developing intelligent decision support systems [14, p. 11–25].

Müller and Novak (2024) addressed the issue of algorithmic bias in automated lending platforms powered by AI. They caution that, in the absence of transparency and robust auditing mechanisms, such systems may intensify inequalities in financial services. Their study advocates for the implementation of ethical review protocols to ensure fairness and accountability [15, p. 70–84].

In summary, while there is broad consensus on the transformative potential of AI in financial services, significant challenges remain, particularly regarding legal frameworks, ethical norms, algorithmic transparency, and governance. The reviewed studies lay a comprehensive theoretical groundwork for this research and underscore the unresolved question of how to construct a financial AI governance model that effectively balances innovation, security, and regulatory compliance.

Despite notable advancements in the study of artificial intelligence (AI) within the financial services sector, much of the existing literature remains concentrated on technical applications—such as case studies and algorithm development—without establishing a comprehensive framework for cross-regional comparison. There is a marked lack of in-depth analysis concerning AI-related financial governance, regulatory approaches, and ethical standards across different countries and regions. Notably, significant divergences in policy priorities, data governance practices, and normative frameworks between the United States, China, and the European Union are frequently oversimplified in current studies. As a result, the real-world implications of these differences on the effectiveness, security, and sustainability of AI applications in finance are insufficiently addressed.

Moreover, present research often overlooks the systemic financial risks and outsourcing challenges associated with AI implementation. Issues such as over-dependence on third-party AI vendors, opacity in algorithmic decision-making ("black box" concerns), and discriminatory practices arising from biased training data pose serious threats to financial system stability, particularly under crisis conditions or extreme market fluctuations. These critical concerns remain underexplored, offering a clear and necessary direction for further investigation.

Accordingly, the development of a holistic analytical framework that integrates technological deployment, risk evaluation, and regulatory governance—while incorporating comparative case studies from diverse national contexts—can bridge current theoretical and empirical gaps. Such a framework would not only enhance academic understanding but also offer valuable insights for shaping more robust and adaptive AI financial regulatory policies.

The primary aim of this study is to conduct a systematic analysis of the implementation pathways, risk factors, and governance structures associated with the use of artificial intelligence (AI) in the financial services sector. Particular emphasis is placed on evaluating the real-world impact of AI on enhancing operational efficiency, improving customer experience, and reinforcing risk management. By undertaking a comparative analysis of the policy orientations, technological applications, and regulatory frameworks in the United States, China, and the European Union, the research seeks to uncover how institutional differences shape the influence of AI on financial system stability, transparency, and regulatory compliance.

The central objectives of this research are as follows:

1. To identify the primary challenges and emerging risks encountered by financial institutions in the adoption of AI technologies.

2. To evaluate the effectiveness and limitations of existing governance models for AI in finance.

3. To develop policy recommendations aimed at enhancing the transparency, resilience, and regulatory responsiveness of AI-driven financial systems.

Drawing upon an integrated methodology that includes literature review, secondary data analysis, and comparative policy assessment, this study aspires to contribute both theoretical insight and practical guidance toward the development of a more secure, efficient, and sustainable AI-powered financial ecosystem.

The scientific novelty of this study is embodied in the following aspects:

1. Introduction of a cross-regional comparative lens: In contrast to previous research that predominantly centers on single-country analyses or focuses narrowly on technical applications, this study adopts a cross-national comparative perspective to examine the distinct trajectories of AI deployment in the financial sectors of the United States, China, and the European Union. It provides a multi-dimensional analysis encompassing policy frameworks, data governance models, ethical standards, and technological implementations.

2. Development of a "technology–risk–regulation" analytical framework: This research establishes a comprehensive model that integrates three critical dimensions: technological application pathways, potential risk typologies, and corresponding regulatory strategies. This triadic framework facilitates a holistic evaluation of AI’s multifaceted impact on the financial system.

3. In-depth exploration of systemic risk and outsourcing governance: Moving beyond the prevailing “efficiency-first” narrative in AI discourse, the study emphasizes less-explored systemic risks introduced by AI, including algorithmic opacity, data-driven biases, ethical concerns, and regulatory mismatches. It further addresses the vulnerabilities associated with dependency on third-party AI providers, proposing multi-dimensional mitigation strategies.

4. Strengthening the empirical foundation of policy recommendations: Grounded in the latest legislative texts, industry white papers, and scholarly findings, this research supplements its analysis with concrete case studies to deliver practical and actionable policy recommendations. These proposals aim to support the development of a more resilient, secure, and governable AI ecosystem within the financial domain.

Drawing on a critical review of the existing literature and the analysis of current policy environments, this study proposes the following research hypotheses:

It is hypothesized that within varying institutional contexts, the effectiveness of AI applications in the financial services sector is largely contingent upon the robustness of governance mechanisms and the adaptability of regulatory frameworks. Specifically, governance models that integrate both technological innovation and ethical oversight are more conducive to the sustainable, secure, and responsible use of AI. In contrast to purely efficiency-driven approaches, such balanced frameworks are more likely to enhance financial system stability and foster greater customer trust.

Furthermore, the study advances the following specific hypotheses:

H1: In the absence of clear legal frameworks and coherent regulatory coordination, the presence of algorithmic bias and the “black box” nature of AI systems may give rise to new forms of systemic financial risk.

H2: The construction of an AI financial governance model grounded in principles of transparency, accountability, and international data cooperation can significantly enhance both the public acceptance and regulatory compliance of AI technologies.

To achieve the stated research objectives and empirically test the proposed hypotheses, this study adopts a multi-method research design that integrates both qualitative and comparative approaches. The methodological framework includes the following components:

1. Literature Analysis: A systematic review of the most recent academic literature (2023–2024) on the application of artificial intelligence in the financial sector, both domestically and internationally, is conducted. This process identifies core theoretical perspectives, prevailing analytical frameworks, and existing academic debates. The findings from this literature review form the theoretical foundation for constructing the analytical framework of the present study.

2. Comparative Analysis: This study selects the United States, China, and the European Union as representative cases for horizontal comparison. Key dimensions of comparison include policy orientation, technology adoption rates, regulatory systems, and ethical standards related to AI in financial services. The objective is to uncover how institutional differences shape the implementation and outcomes of AI applications in financial practice.

3. Regulatory Analysis: To address the legal and regulatory challenges associated with AI in financial services, this study undertakes a detailed examination of pivotal regulatory documents. These include China’s New Generation Artificial Intelligence Development Plan, the California Consumer Privacy Act (USA), and the European Union’s AI Act. The analysis focuses on how these legal frameworks influence the compliance, transparency, and accountability of AI systems in the financial domain.

4. Case Study Analysis: Drawing from the real-world practices of leading fintech companies—such as JPMorgan Chase, Ant Group, and ING—this study investigates the deployment of AI technologies in areas including financial management, customer service, and algorithm governance. The case studies highlight both the practical benefits and potential governance challenges associated with AI integration in financial institutions.

In addition to these core methods, the study also incorporates industry reports, regulatory white papers, and official policy texts to ensure strong practical relevance and alignment with current policy discourse. By synthesizing these diverse research methods, the study aims to effectively bridge theoretical inquiry with real-world observations, thereby enhancing the systematic rigor, logical coherence, and explanatory power of the analysis.

Artificial intelligence (AI) has emerged as a transformative force in financial services, projected to contribute $1 trillion annually to global economic growth by 2030. This evolution, driven by advancements in big data and machine learning, is reshaping how institutions manage risk, enhance customer experiences, and optimize operations. Researchers agree that now there is no single common definition of artificial intelligence.

Research on artificial intelligence began in the 1950s, when Alan Turing in his groundbreaking article Computing Machinery and Intelligence tried to answer the question of whether machines can think. At that time, he applied a comprehensive approach to the problem posed on the logical, philosophical and technical levels. He was also the creator of a test that allowed to check the intelligence of machines (later called the Turing test). In 1946. Turing formulated the assumption that "in 30 years, asking a question to a computer will be as easy as asking a question to a human". It has been 77 years since these words were uttered, and yet they are still not 100% confirmed.

In the face of the rapidly advancing digital revolution, the financial sector is among the fields in which artificial intelligence is implemented and used most intensively [16, p. 177]. This information also appears in one of the European Commission's press releases, where, in the context of measures to develop the financial market, it is stated that the financial sector is "the largest user of digital technologies, which is the main driver of the digital transformation of the economy and society" [17, p. 269].

In the last decade, a trend has been observed confirming the rapid growth of artificial intelligence, manifested primarily in a decrease in the costs associated with data processing and use. These premises are confirmed by both the greater availability and the number of data analyzed [18, p. 1052]. In addition, expenditure on improving IT infrastructure is also characterized by an upward trend [19, p. 448].

Artificial intelligence allows entrepreneurs to make decisions faster and more accurately. In the area of finance, it can help with risk management, fraud detection, customer service or compliance with regulations. In addition, regulatory reporting and financial planning requirements can contribute to the use of artificial intelligence in improving business processes. By gaining better insight into customer behavior, as well as developing new, innovative products or services, companies can increase their level of competitiveness.

Selected examples of the use of artificial intelligence in finance

Artificial intelligence in the financial sector is used quite widely for both internal and external needs. Numerous cases of its use include:

1. Risk assessment,

2. Fraud detection,

3. Financial advisory services,

4. Investment decisions in finance,

5. Credit decisions,

6. Private banking,

7. General automation of processes in finance,

8. Compliance with regulations and rules.

One of the basic areas of application of artificial intelligence in the area of finance is risk assessment. It is a key element in the operation of any financial institution, especially those that lend funds or invest in securities. Artificial intelligence helps companies predict and assess risk, based on the vast amount of available data. Currently, there is a growing ability of banks and other financial institutions to make more accurate decisions about the creditworthiness of potential borrowers. In addition, these activities are intended to reduce the likelihood of granting loans or credits for improper purposes or to those groups of borrowers who do not intend to repay their liabilities. An example is the operation of BIK (Credit Information Bureau), which, using the potential of the credit and loan customer base, undertook to carry out the study in question. It consisted in determining the predictive effectiveness of credit scoring created using the machine learning method, and then comparing it with the results obtained thanks to traditional statistical methods that have been used for years. For this m.in purpose, the XAI (explainable artificial intelligence) method, the Gini coefficient (a measure of the correctness of predictions) and the gradient boosting machine technique were used. As a result, it was found that the machine learning methods used have a significant impact on increasing the efficiency of determining the level of credit risk and, at the same time, increasing the profits of the financial institution [20, p. 7].

Another area of application of AI in the field of finance is fraud detection. By analyzing user behavior and comparing it to the pattern of regular purchases, sales, and commerce, AI can detect fraud by generating alerts when certain behaviors deviate from expectations. In this case, machine learning is used, m.in, for example, in anti-fraud systems or anti-money laundering and counter-terrorism financing systems. An IBM study predicts that global losses from this type of fraud will reach $44 billion by 2025, and 72% of enterprise leaders will consider this result a significant concern [21, p. 1314]. In addition, artificial intelligence is used in the financial sector for compliance management, i.e. in compliance with applicable laws and regulations.

Artificial intelligence in the field of finance is also used in financial advisory services. It solves the problems associated with processing huge amounts of data and provides more specific financial advice for people who do not have sufficient knowledge and ability to carry out such analyses. For example, let's assume that a client has an amount of 500,000 USD invested in investment funds and wants to know which funds will achieve good results and which will not in each time perspective (several weeks, months of years or longer). The use of artificial intelligence in this area allows you to receive a personalized report on the performance of funds in each period, and as a result, a chance to make more adequate decisions from the perspective of the investment. An AI-based tool for verifying contractors is, for example, the Temida system, which makes it possible to analyze over 3000 factors affecting the solvency of a given company within 30 seconds and calculate the probability of its debt. By using this intelligent data-driven verification, we will be able to quickly and easily eliminate dishonest contractors even before we start working with them.

Another area of application of artificial intelligence is investment decisions in finance (trading). Stock markets are influenced by many factors, which is why they sometimes surprise even the most experienced investors. In this aspect, artificial intelligence systems learn through experience and thus lead to more accurate predictions. Compared to humans, artificial intelligence, in addition to being able to analyze huge data sets, can make faster predictions that consider learning from mistakes, such as predicting a decline in the bitcoin exchange rate. Artificial intelligence can also take advantage of possible errors in its models, which allows it to provide much more accurate data in subsequent applications. According to JP Morgan, in 2020, the feedback obtained using algorithms reached USD 10 million and covered 60% of all transactions, and experts predict that the value of these operations will grow at a dizzying pace [22, p. 43].

The impact of artificial intelligence is also visible in credit decision-making. The analysis of customers' credit history by artificial intelligence is based on the activity on the bank account, which in turn translates into risk assessment in granting loans. According to Forbes magazine, the use of machine learning used to predict incoming cash flow or credit scores applies to 70% of financial institutions [23, p. 465]. Technology allows you to determine the factors (such as income, age, or type of property you want to buy) that you should consider when borrowing money. For example, the lender may reject the application if the customer is interested in buying an expensive house but has a bad credit history. AI could also be used to analyze many different data points that are related to an individual applicant before deciding whether to qualify them for a loan.

Artificial intelligence is also used in private banking. Customer service with the use of chatbots (the equivalent of waiting in line for a phone call) has recently been gaining popularity. Artificial intelligence can now be used to provide highly individualized financial advice using information on user activity and various data collected from other non-banking sources.

Also, the general automation of processes in finance is possible thanks to the use of artificial intelligence. Repetitive (e.g. simple mathematical operations) or more complex tasks (financial modelling) are subject to automation. This process, if well programmed, can replace financial specialists at work, who could focus on other activities at the same time. It should be remembered that some tasks may require human intuition, creative approach or emotional intelligence.

The use of AI should comply with regulations and applicable rules. However, all regulations in this area remain complicated and difficult to understand in many issues, especially when it comes to cross-border transactions (whether, for example, the sale of a bond issued in Europe and purchased by an investor in China can be resold in the US – in this case, several questions arise: where to get information from, whether such a transaction is legal, whether different rules apply in each country).

It is desirable to apply artificial intelligence in this area through the rapid and accurate analysis of countless amounts of information. In an analogous situation, a person will not give an answer in such a short time. Thanks to artificial intelligence, it is possible to analyze the financial statements of a given company to check whether they have been prepared in accordance with the applicable regulations, which also makes it possible to catch any discrepancies before they grow into serious problems.

I would like to emphasize that the above-mentioned applications are exemplary, and their list is incomplete, as a lot is still changing in this matter, considering the development of the technology itself and the expectations of financial institutions and their customers.

Comparative Analysis of AI Adoption in Financial Services: United States, China, and the European Union

The adoption of artificial intelligence (AI) in financial services varies significantly across regions, influenced by unique regulatory frameworks, market dynamics, and technological priorities. This section analyzes the United States, China, and the European Union (EU), highlighting their approaches to integrating AI in financial services and the resulting implications.

1. United States: A Hotbed for Innovation and Risk Management:

1. Investment in AI: According to CB Insights, U.S.-based fintech companies raised $23 billion in 2022, with 38% of the funding allocated to AI-driven innovations.

2. Efficiency Gains: JPMorgan’s AI-driven COiN platform reduces contract review time by over 95%, saving an estimated $1.3 billion annually in operational costs.

Fig. 1 U.S. AI Investment in Financial Services (2020-2022) [8]

The United States has established itself as a global leader in AI-driven financial services, largely due to its dynamic fintech ecosystem and relatively relaxed regulatory environment. Companies such as Stripe, Robinhood, and PayPal leverage AI for personalized financial solutions, real-time fraud detection, and predictive analytics. Major banks, including JPMorgan and Bank of America, utilize AI to automate risk assessment, saving billions annually.

One notable example is JPMorgan’s "COiN" (Contract Intelligence) platform, which processes and analyzes 12,000 legal documents in mere seconds, a task that would require approximately 360,000 human work hours. Additionally, AI-powered chatbots like Bank of America’s “Erica” enhance customer service, handling over 100 million user interactions since its launch.

Regulatory frameworks like the California Consumer Privacy Act (CCPA) allow innovation while safeguarding consumer rights. However, the decentralized regulatory landscape occasionally creates inconsistencies across states, posing challenges for nationwide AI implementation.

2. China: Government-Led Innovation for Scale:

1. AI Adoption Rate: Over 90% of Chinese financial institutions have integrated AI solutions, particularly in mobile payments and credit scoring.

2. Mobile Payments Market: Platforms like Alipay and WeChat Pay process over 200 billion transactions annually, with 70% leveraging AI for fraud detection.

3. Regulatory Support: The "AI Development Plan" allocated over $30 billion in funding between 2017 and 2022 to accelerate AI adoption in financial markets.

China's financial sector has become a global AI hub, driven by state-backed initiatives and a consumer base that readily adopts digital technologies. Platforms like Alipay and WeChat Pay have transformed everyday transactions by incorporating AI for fraud detection and financial planning. The integration of facial recognition and behavioral biometrics in credit scoring is another hallmark of China’s AI adoption.

The "Next Generation Artificial Intelligence Development Plan," launched by the Chinese government in 2017, aims to make China a world leader in AI by 2030. This top-down approach accelerates AI integration but raises concerns about data privacy and ethical considerations, as regulatory oversight prioritizes technological progress over consumer protections.

China’s Ant Group exemplifies the country’s AI-driven financial landscape, using algorithms to assess creditworthiness for micro-loans. This approach has enabled financial inclusion for millions previously excluded from traditional banking, but it has also faced scrutiny, such as during the halting of Ant’s IPO in 2020 over regulatory concerns.

3. European Union: The Advocate of Responsible AI:

1. Ethical AI Compliance: 68% of European financial institutions report prioritizing GDPR-compliant AI systems over cost efficiency.

2. AI in Compliance: ING reduced false positives in transaction monitoring by 40% using AI-driven compliance tools.

3. AI Funding: The EU lags the U.S. and China, with $7.5 billion invested in AI technologies in 2022, primarily targeting ethical AI frameworks.

Fig. 2 AI Investment in Financial Services (2022) [13]

The EU takes a more cautious approach to AI adoption, emphasizing transparency, accountability, and ethical considerations. The General Data Protection Regulation (GDPR) is a cornerstone of AI regulation, mandating stringent data privacy standards that shape AI applications in finance.

European financial institutions such as ING and Revolut focus on compliance and consumer trust. ING employs AI to monitor transactions for regulatory compliance and fraud detection, ensuring adherence to the EU’s strict legal standards. Meanwhile, Revolut uses machine learning to offer highly personalized financial services while maintaining robust security.

Although these regulations foster trust and long-term sustainability, they can slow innovation compared to the U.S. and China. The EU’s emphasis on ethical AI provides a competitive edge in establishing responsible and globally accepted AI practices.

The United States, China, and the EU represent distinct models of AI adoption in financial services. While the U.S. leads in innovation, China excels in scalability, and the EU sets the standard for responsible AI practices. Together, these approaches shape the global AI landscape, balancing the benefits of technological advancement with the need for ethical governance.

Analytical Insights into AI Adoption in Financial Services

1. Regulatory Impact and Market Dynamics. The regulatory environment is a defining factor in how countries adopt AI in financial services. The U.S.’s relatively lenient and fragmented regulatory framework allows rapid innovation, particularly among fintech startups. However, this creates risks, such as inconsistent compliance standards across states. For example, while the California Consumer Privacy Act (CCPA) provides robust data protection, states like Texas have more relaxed rules, leading to uneven AI application.

China, in contrast, benefits from centralized planning under initiatives like the "AI Development Plan." This enables rapid AI deployment, particularly in state-backed financial services, where adoption is nearly universal. However, the lack of stringent privacy laws raises ethical concerns, especially in credit scoring systems that utilize personal data extensively.

In the EU, GDPR ensures strict data protection, fostering consumer trust but often at the expense of speed in innovation. For instance, AI applications in fraud detection in the EU must undergo rigorous testing to ensure compliance, which delays deployment but ensures higher reliability and consumer confidence.

2. Socioeconomic Factors Influencing Adoption. The socioeconomic environment also plays a critical role in shaping AI adoption:

  • United States: High competition in the financial services sector drives innovation. Institutions prioritize operational efficiency, leveraging AI to automate risk assessment and compliance processes, reducing labor costs by up to 40%.
  • China: A mobile-first economy with a high rate of digital payment adoption allows AI integration to scale rapidly. Alipay and WeChat Pay not only dominate domestic transactions but are also expanding into international markets, raising questions about data sovereignty.
  • European Union: Higher public scrutiny and ethical concerns lead to cautious AI integration. Initiatives focus on reducing systemic risks, such as algorithmic bias, ensuring that financial products are accessible without discriminatory practices.
  • 3. Technological Ecosystems and Resource Allocation. Access to cutting-edge technology and resources further differentiates these regions:

  • United States: A well-established AI ecosystem, supported by Silicon Valley and major tech players like Google and IBM, enables financial firms to deploy sophisticated machine learning models.
  • China: Heavy government investment in AI infrastructure, including cloud computing and big data analytics, supports large-scale financial experiments. For instance, the People’s Bank of China uses AI to monitor systemic risks in real time.
  • European Union: Investment is directed toward responsible AI research, often in collaboration with academic institutions. This ensures sustainable development, but limits scalability compared to China or the U.S.
  • Opportunities for the use of artificial intelligence

    The development of new technologies affects various areas, including the financial market. Entities operating on it, with the help of artificial intelligence techniques, create offers that are best suited to customer expectations. AI can improve productivity, efficiency, and accuracy with minimal human effort, and this makes it extremely efficient. The use of artificial intelligence in the financial market sector allows you to achieve the following benefits:

    1. Positive impact on the image,

    2. Reduction of the level of costs,

    3. Increase in productivity and speed of work,

    4. Increasing access to financial services,

    5. Improving decision-making processes,

    6. Better identification of threats,

    7. Improve internal control.

    Companies with a strong focus on innovation and modernity use artificial intelligence using new technologies. The solutions offered by artificial intelligence have a significant impact on the reputation of financial institutions and their perception by their target audience, for which image plays an important role in many cases.

    To compare the level of costs, the speed and efficiency of human work were compared with the capabilities of artificial intelligence. The technology of the American bank JPMorgan Chase & Co., based on artificial intelligence, is able to review about 12 thousand documents in just a few seconds, while a human would have to spend about 360 thousand hours on this activity [24, p. 37]. Translating this directly into the labor market, we can talk about a significant reduction in the costs associated with hiring employees thanks to such an improvement in the business process.

    Processing multidimensional data also contributes to increased productivity, which, compared to human workload, requires longer personalization of financial offers with a higher risk of error, as opposed to when data is processed using an artificial intelligence algorithm. As for the waiting time, it is also shortened even in the case of a specific offer, an example of which can be a credit decision – the data considered by the employee is given to a multi-stage formal identification [25, p. 143]. Thanks to the use of artificial intelligence, such a decision will be made with greater precision and much faster.

    Increasing access to financial services using artificial intelligence translates into the development of highly personalized products and services, which gives the opportunity to offer customers financial services that are better suited to their needs [26, p. 180]. Better insight into customer preferences and behavior patterns is extremely important in the context of the application of artificial intelligence methods in the financial services sector, as they are used to design better, more tailored and personalized products, services or offers. The process of automating customer service obtained using artificial intelligence algorithms translates into better adjustment of a given service to the individual needs of the consumer or company, and even – looking more broadly – can support the financial awareness of the society, which may prove particularly useful for people who are not professionally related to finance. An example of this type of solution are tools that facilitate currency conversion or loan installments.

    Analysts can apply artificial intelligence, thanks to which decision-making processes are improved by making faster choices based on data analysis. Automating tasks allows analysts to save time that they can later spend on other types of work, for example, supporting clients in the application process or conducting financial modeling.

    Better identification of threats is another area of application of artificial intelligence in the financial sector. The potential of technology is widely used by financial institutions to reduce fraud and extortion [27, p. 389]. Detecting fraud and irregularities is one of the most frequently cited reasons for the adoption of artificial intelligence solutions by financial service providers [28, p. 17]. It is also worth noting that, unlike traditional algorithms, machine learning systems contribute to the search for new, previously unknown types of abuse [29, p. 321]. Thanks to the use of artificial intelligence algorithms in this field, financial institutions can analyze transactions that seem suspicious in advance and apply mechanisms to prevent money laundering, as well as the financing of crime [30, p. 113].

    Audit plays a key role in finance. By improving internal control in this area, the company can significantly reduce e.g. internal fraud by monitoring account activity non-stop.

    The above-mentioned opportunities and benefits resulting from the use of artificial intelligence are an exemplary set of solutions, therefore they should not be treated on an exclusive basis. With the development of technology, there will be further attempts to functionalize artificial intelligence within the financial market. It is worth noting that the use of artificial intelligence algorithms in the field of finance is no longer only a matter of choosing an action strategy, but also a condition for development. In addition, the measurable benefits that come from the implementation of artificial intelligence can significantly contribute to achieving m.in long-term market competitiveness [31, p. 1235].

    Just as important as the benefits of using artificial intelligence are the risks associated with this phenomenon. Due to the short research perspective, many of the potentially negative consequences for the financial sector have not yet been identified. However, there is awareness indicating the existence of certain barriers and gaps in the development of artificial intelligence. Potential threats include lack of clear and transparent guidelines for the use of artificial intelligence in the financial sector, lack of adaptation to reality, the threat of cybercrime and the lack of solutions in this area, a significant increase in costs, the risk of failure, legal uncertainty related to overregulation, the lack of appropriate organizational and technical solutions.

    While the benefits of AI in financial services are undeniable, its rapid adoption also raises several concerns:

    1. Algorithmic Bias: AI-driven credit scoring models can inadvertently amplify biases, leading to discriminatory lending practices.

    2. Cybersecurity Threats: Increasing reliance on AI exposes financial institutions to sophisticated cyberattacks. For example, adversarial attacks can manipulate AI models to produce false predictions.

    3. Systemic Risks: The overuse of AI in high-frequency trading has already resulted in market anomalies, such as flash crashes, highlighting the need for robust monitoring systems.

    Despite these challenges, ongoing advancements in AI governance and technology continue to address these issues, paving the way for more reliable applications in finance.

    The constantly changing reality and the development of technology require the adaptation of individual tools to the changing reality. As noted, some of the artificial intelligence techniques have not been tested in the conditions of financial crises [32, p. 79], which makes machine learning models based on historical data that does not correspond in any way to the current situation. As a result, the resulting models and algorithms may reflect reality to an inadequate extent [33, p. 14]. From the point of view of human values, decisions made on the basis of algorithms, despite their decisive advantage in terms of working time, may be useless [34, p. 19]. Also under the law, there is a risk of privacy protection or liability in the event of losses that may be caused using artificial intelligence [35, p. 492].

    Artificial intelligence technology certainly brings new challenges, primarily in terms of data privacy and security. The threat of cybercrime is visible, while there are no sufficient solutions in this area. In the context of financial markets, the stability of not only individual institutions, but also the entire sector depends on cybersecurity. As the volume of digital transactions continues to grow, which is largely conducive to attacks, financial market entities are using newer and newer security measures to be able to counteract them. However, hackers are aware that as automation increases for repetitive tasks, so does the opportunity for cyberattacks, as the number of potential vulnerabilities also increases [36, p. 500]. Due to the introduction of false data into algorithmic models, software is also vulnerable to this type of attack, which can consequently lead to phishing [37, p. 452].

    The area of significant cost growth can be approached in a multifaceted way with respect to financial market entities, as it includes expenditure on implementation (purchase or creation of software and its subsequent dedicated maintenance), development (adaptation to emerging changes), maintenance of artificial intelligence, costs of possible errors or other failures, and even social costs (automation of processes, including reduction in the number of jobs as a result of human replacement by algorithms or robots). In addition, the most innovative technologies may be available in principle only to large entities, mainly due to the capital needed to invest in them [38].

    The potential risk of failure, on the other hand, makes it necessary to pay attention to cases in which financial institutions have implemented algorithms that operate in an unexpected way, which has consequently led to errors and sudden market crashes [39, p. 350]. Such failures are usually spectacular and therefore costly, and they also have a huge impact on the stability of financial markets both now and in the future [40, p. 9]. In addition, systemic limitations and barriers sometimes go far beyond the subject of artificial intelligence.

    Conclusion

    The currently observed changes in the financial market concern the dynamic development of artificial intelligence, which, due to its potential, is becoming an area of interest for many entities on both a national and global scale. The use of artificial intelligence gives financial organizations the ability to make better decisions by analyzing and evaluating the vast amount of data processed in real time and the ability to derive information not only from domestic but also from global financial markets. An extremely important advantage is the fact that both data collection, calculation and analysis are carried out simultaneously.

    Finance in its broadest sense is a human endeavor, but thanks to the use of systems based on artificial intelligence, they are becoming more efficient and precise than ever before, as well as more accessible to those interested. Artificial intelligence is revolutionizing the financial sector in a significant way, helping to automate and optimize processes such as credit scoring and risk management. Companies use artificial intelligence, e.g. through machine learning, to reduce their financial risk. Users of artificial intelligence systems – in this case, analysts, accountants, treasurers or investors – can detect fraud more easily and find various types of anomalies, which ensures continuous development of the company.

    The dynamic development of artificial intelligence and its widespread use make market regulations necessary to ensure that the use of solutions created and based on artificial intelligence remains safe, and the results of analyses carried out using it are thus as reliable as possible. Along with technological progress, many legal, economic, social and moral challenges arise for financial market entities using artificial intelligence. It is an indisputable observation that financial markets are facing a huge opportunity to improve many processes and operations thanks to artificial intelligence, which in turn can be a source of many measurable benefits not only for interested entities, but also for the entire economy. Just like about opportunities, you should also remember about threats that require thoughtful action in the area in question.

    The recommendations indicated in the article should make it possible in several cases to look at regulatory constraints that may contribute to both the application and development of current solutions in the field of artificial intelligence. As shown during the discussions, the financial sector is increasingly adopting artificial intelligence systems and is simultaneously facing numerous challenges related to many internal and external processes. The recommendations indicated concern the provision of effective solutions in the field of stability, security and transparency of the use of artificial intelligence in the financial sector. As it has been proven, the increasing use of artificial intelligence entails numerous cases of risk, which are associated with the outsourcing of certain activities to third parties. An additional aspect is digital resilience, which is an important area from the perspective of ensuring the security and stability of the sector. The new regulations prepared in this connection should be of key importance to ensure financial stability in the broadest sense.

    The integration of AI in financial services marks a pivotal moment in the industry’s evolution. To fully harness its potential, institutions must navigate challenges in regulation, ethics, and technology. As AI continues to mature, its role in fostering financial inclusion, enhancing decision-making, and driving global competitiveness will only grow.

    This study is an attempt to collect issues in the form of barriers and challenges and to show the areas that require the immediate creation of new regulations or the updating of the current ones. The proposed recommendations allow us to see both certain limitations and set the direction of the demand for new regulations and solutions. They also point out that the introduction of regulations is an extremely important, desirable and at the same time very complex process.


    Источники:

    1. Zhao Y., Li J., Wang H. Machine Learning Applications in Bank Credit Scoring: Opportunities and Limitations // Journal of Banking Analytics. – 2023. – № 15(1). – p. 24–36. – doi: 10.1016/j.jba.2023.01.003.
    2. Chen R., Liu S. AI-Based Portfolio Optimization under Market Uncertainty // Finance and Technology Review. – 2023. – № 12(2). – p. 78–89. – doi: 10.1080/fintech.2023.12.078.
    3. Wang M., Zhang K., Feng T. Fraud Detection in Mobile Payments Using Deep Learning Techniques // Journal of Financial Risk Management. – 2024. – № 9(1). – p. 55–70. – doi: 10.1108/JFRM-2024-011.
    4. Smith A., Taylor B. AI in Compliance and AML: A Case Study of EU Banks // European Journal of Finance and Regulation. – 2023. – № 17(4). – p. 103–118. – doi: 10.1016/ejfr.2023.04.005.
    5. Lee J., Kim D. Algorithmic Bias in AI-Based Credit Scoring Models // Journal of Ethical Finance Technology. – 2023. – № 6(3). – p. 15–32.
    6. Nguyen L. T., Müller K., Rossi G. GDPR-Compliant AI in Financial Services: A Regulatory Analysis // Data Protection Law Review. – 2024. – № 11(2). – p. 41–56. – doi: 10.1016/dplr.2024.02.001.
    7. Huang X., Wang J., Dupont L. Comparing AI Financial Strategies: U.S., China, and the EU // International Journal of Fintech Studies. – 2024. – № 5(1). – p. 60–74. – doi: 10.1108/IJFS.2024.01.006.
    8. Garcia M., Alvarez P., Lu Y. Conversational AI in Finance: Enhancing Customer Experience // AI in Service Innovation. – 2023. – № 3(3). – p. 88–99. – doi: 10.4018/AISI.2023.033007.
    9. Kumar N., Singh R. AI-Driven Microfinance Credit Scoring in South Asia // Journal of Inclusive Finance. – 2023. – № 7(2). – p. 22–38. – doi: 10.1080/JIF.2023.072002.
    10. Jin Y., Zhang H. AI and Systemic Risk in Financial Markets: An Empirical Warning // Global Finance Review. – 2023. – № 13(4). – p. 127–140. – doi: 10.1080/GFR.2023.134001.
    11. Xu L., Zhao M., Chen H. Enhancing Risk Management in Local Banks through AI // Asia-Pacific Journal of Financial Innovation. – 2024. – № 8(1). – p. 99–115. – doi: 10.1108/APJFI.2024.08.011.
    12. Park S., Lee C. Financial Chatbots and Human-AI Interaction in Banking // Journal of AI & Human Behavior. – 2023. – № 4(2). – p. 51–66. – doi: 10.1016/JAHB.2023.02005.
    13. Torres D., Salazar R. Blockchain and AI in Regulatory Reporting // Technology & Regulation Review. – 2023. – № 5(3). – p. 118–132. – doi: 10.1108/TRR.2023.053007.
    14. Zhao K., Wen T. Financial Decision-Making Optimization Based on AI Algorithms // Journal of Intelligent Economic Systems. – 2024. – № 10(1). – p. 11–25. – doi: 10.1016/JIES.2024.01001.
    15. Müller F., Novak S. Risk of AI-Induced Bias in Automated Lending Platforms // Ethics & Digital Finance. – 2024. – № 6(1). – p. 70–84. – doi: 10.1016/EDF.2024.06008.
    16. Akhtar P., Khan Z., Frynas J. G., Tse Y. K., Rao-Nicholson R. Enablers and barriers of artificial intelligence adoption in banking: An analysis using Porter’s diamond model // Journal of Business Research. – 2019. – № 101. – p. 170-182. – doi: 10.1016/j.jbusres.2019.03.022.
    17. Huang Y., Verma R. Artificial intelligence applications in financial services // Journal of Financial Services Marketing. – 2019. – № 24(4). – p. 263-277. – doi: 10.1057/s41264-019-00065-9.
    18. Kamble S. S., Gunasekaran A., Sharma R. Analysis of the driving and dependence power of barriers to adopt artificial intelligence in the banking sector // International Journal of Production Research. – 2020. – № 58(3). – p. 1043-1063. – doi: 10.1080/00207543.2019.1595430.
    19. Chitra K., Subashini B. Artificial intelligence in banking: Challenges and opportunities // Procedia Computer Science. – 2019. – № 165. – p. 446-453. – doi: 10.1016/j.procs.2020.01.038.
    20. Nguyen T. T., Sinha A. P. Impact of artificial intelligence on business performance in the banking sector // Journal of Information Technology. – 2020. – № 35(1). – p. 3-14. – doi: 10.1177/0268396220911927.
    21. Gorgoglione M., Panniello U. Beyond customer experience models: Identifying idiosyncratic perceptions // International Journal of Bank Marketing. – 2018. – № 36(7). – p. 1311-1328. – doi: 10.1108/IJBM-11-2017-0240.
    22. Durai T., Stella G. Role of artificial intelligence in financial services industry // International Journal of Management Studies. – 2019. – № 6(3). – p. 40-47. – doi: 10.18843/ijms/v6i3/04.
    23. Komulainen H., Saraniemi S. Customer centricity in mobile banking: A customer experience perspective // International Journal of Bank Marketing. – 2019. – № 37(2). – p. 456-472. – doi: 10.1108/IJBM-04-2018-0096.
    24. Lee K. C., Shin D. H. Fintech: Ecosystem, business models, investment decisions, and challenges // Business Horizons. – 2020. – № 63(1). – p. 35-45. – doi: 10.1016/j.bushor.2019.09.003.
    25. Verma P., Sehgal S. Leveraging artificial intelligence for enhancing customer experience and efficiency in the banking industry. In Bhardwaj, B., et al. (Eds.), AI and Emotional Intelligence for Modern Business Management. IGI Global. , 2020. – 140-157 p.
    26. Gao L., Melero-Polo I., Sese F. J. Customer equity drivers, customer experience quality, and customer profitability in banking services: The moderating role of social influence // Journal of Service Research. – 2020. – № 23(2). – p. 174-193. – doi: 10.1177/1094670519889358.
    27. Arner D. W., Barberis J., Buckley R. P. FinTech, RegTech, and the reconceptualization of financial regulation // Northwestern Journal of International Law & Business. – 2017. – № 37(3). – p. 371-413. – doi: 10.2139/ssrn.2949029.
    28. Mohapatra S. Artificial intelligence and the financial services industry: Challenges and opportunities // Journal of Business Strategy. – 2020. – № 41(4). – p. 15-21. – doi: 10.1108/JBS-10-2019-0202.
    29. Gronroos C. Adopting a service logic for marketing // Marketing Theory. – 2006. – № 6(3). – p. 317-333. – doi: 10.1177/1470593106066794.
    30. Cojoianu T., Ascui F., Clark G. L., Hoepner A. G., Wojcik D. Does responsible investment pay? Analyzing the financial performance of green bonds // Journal of Banking & Finance. – 2020. – № 113. – p. 105782. – doi: 10.1016/j.jbankfin.2019.105782.
    31. Balakrishnan V., Das D. Impact of artificial intelligence on the banking sector in Malaysia // International Journal of Emerging Markets. – 2020. – № 15(6). – p. 1221-1241. – doi: 10.1108/IJOEM-01-2019-0005.
    32. Siau K., Wang W. Artificial intelligence (AI) ethics: Ethics of AI and ethical AI // Journal of Database Management. – 2020. – № 31(2). – p. 74-87. – doi: 10.4018/JDM.2020040105.
    33. Jaiswal M. P., Kumar R. Artificial intelligence in the banking sector: Application and challenges // Journal of Enterprise Information Management. – 2021. – № 34(1). – p. 1-22. – doi: 10.1108/JEIM-06-2019-0171.
    34. Srinivasan A., Arunasalam B. Leveraging big data analytics to reduce healthcare costs // IT Professional. – 2019. – № 21(1). – p. 17-23. – doi: 10.1109/MITP.2018.2881327.
    35. Lee I., Shin Y. J. Machine learning and AI in marketing – Connecting computing power to human insights // International Journal of Research in Marketing. – 2020. – № 37(3). – p. 481-501. – doi: 10.1016/j.ijresmar.2020.04.005.
    36. Oliveira T., Thomas M., Espadanal M. Assessing the determinants of cloud computing adoption: An analysis of the manufacturing and services sectors // Information & Management. – 2014. – № 51(5). – p. 497-510. – doi: 10.1016/j.im.2014.03.006.
    37. Loureiro S. M. C., Guerreiro J., Tussyadiah, I. Artificial intelligence in business: State of the art and future research agenda // Journal of Business Research. – 2020. – № 123. – p. 448-463. – doi: 10.1016/j.jbusres.2020.09.068.
    38. Shankar V. How artificial intelligence (AI) is reshaping retailing // Journal of Retailing. – 2018. – № 94(4). – p. vi-xi. – doi: 10.1016/j.jretai.2018.10.005.
    39. Ding X. D., Huang Y., Verma R. Customer experience in online financial services // Journal of Service Management. – 2011. – № 22(3). – p. 344-366. – doi: 10.1108/09564231111136895.
    40. Ransbotham S., Kiron D., Prentice P. K. // MIT Sloan Management Review. – 2016. – № 57(2). – p. 1-13. – url: https://sloanreview.mit.edu/projects/analytics-talent-dividend/.

    Страница обновлена: 17.04.2025 в 10:22:45