IJBMS, 2025 Page No.01-30
IJBMS
1
Volume05, Issue 03, 2025,
Publish Date: 10-03-2025
DOI: -
https://doi.org/10.55640/ijbms-05-03-01
AI IN BUSINESS ANALYTICS FOR FINANCIAL RISK
ASSESSMENT: SURVEY INSIGHTS FROM THE BANKING
AND INSURANCE INDUSTRIES
Sonia Akter
Mercy University, USA
Tanmoy Saha Turja
Mercy University, USA
Amjad Hossain
Mercy University, USA
Sanjida Alam Eshra
Trine University, USA
Iftekhar Rasul
St. Francis College, USA
INTERNATIONAL JOURNAL OF BUSINESS AND MANAGEMENT SCIENCES (Open access)
ABSTRACT
In the United States, artificial intelligence (AI) has become a transformative force in the business
analytics area related to financial risk assessment for banking and insurance industries. The aim of this
research is to assess adoption, effectiveness and challenges of AI driven risk assessment models, by
analyzing data collected through a survey, which was distributed to 200 financial professionals across
the U.S. According to the findings, AI plays an important role in increasing the accuracy of fraud
detection, reducing credit risk, predicting market risk, minimizing operational risk and other decisions
and optimizing cost efficiency at the financial institutions. The adoption of AI technology in improving
the efficiency of the pharmaceutical industry is hindered by some key barriers such as concerns about
data privacy, compliance regulations, high implementation costs and shortage of AI specialists.
According to the results, financial institutions need to expand governance frameworks to ensure the
regulatory alignment and ethics in using AI in a transparent way while maintaining safe risk assessment
model. The contribution of this study to the current debates on AI and finance risk management, as well
as implications for both the policymakers and financial industry practitioners, might include practical
advice and recommendations to financial institutions and researchers on better integrating AI in banking
and insurance risk assessment systems.
Artificial intelligence, business analytics, financial risk assessment, banking, insurance, fraud
detection, credit risk, market risk forecasting, operational efficiency, regulatory compliance, data
privacy, AI governance, machine learning, predictive analytics, U.S. financial institutions.
IJBMS, 2025 Page No.01-30
IJBMS
2
© The Author(s) 2025.
Open Access
This article is licensed under a Creative Commons Attribution 4.0 International License,
which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if
changes were made. The images or other third-party material in this article are included in the
article’s
Creative Commons
license, unless indicated otherwise in a credit line to the material. If material is not included in the
article’s
Creative
Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a copy of this license, visit
http://creativecommons.org/licenses/by/4.0/
INTRODUCTION
In the United States, banking and insurance industries are utilizing its use of artificial intelligence (AI) as
a transformative force in business analytics for financial risk assessment. Recent advancements in the
financial markets, imposing enhanced regulatory requirements, ever increasing cybersecurity threats,
has encouraged the financial institutions to use AI based predictive analytics, machine learning models
and big data technologies to improve risk management and decision-making process (Paul, Sadath,
Madana, 2021; Ahmadi, 2024). Vast financial data could be processed in real time by AI not so long ago,
helping financial institutions to detect fraud accurately, credit risk and reduce operational risk to
maintain competitive advantage in managing financial uncertainties (Aziz & Andriansyah, 2023). While
AI adoption bears several benefits, the usage of AI is not spreading equally in U.S. financial institutions,
mainly due to ongoing worries about data privacy, regulatory compliance and costs of implementation
as well as training of workforce (Herrmann & Masawi, 2022; Nwaimo, Adewumi, & Ajiga, 2022).
In the U.S, the banking sector is gradually shifting toward the use of AI business intelligence for the
purposes of enhanced credit underwriting, improved fraudulent transactions detection and for
improved prediction of loan defaults. Traditional credit scoring methods utilize historical financial data
and some machine learning algorithms to generate models that reduce errors in scoring clients and
increases the effectiveness of identifying high risk clientele (Bello, 2023; Islam et al, 2024). AI helps in
discovering fraudulent transactions as pattern recognition and anomaly detection by the machine
learning algorithms help in detecting the fraudulent activities in real time (Chowdhury et al, 2024;
Pattnaik, Ray, & Raman, 2024). AI integration is also being used by the insurance industry, mainly in
claims processing, risk pricing and fraud detection; here, AI model aids the insurers in faster assessment
of the risks they need to cover when dealing with policy holders (Kannan, 2024; Aleksandrova et al,
2023). These applications are showing how the trend of artificial intelligence’s more substantial
contribution to financial risk assessment helps U.S. financial institutions increase their risk visibility,
improve their compliance and make more accurate decisions (Rahmani & Zohuri, 2023).
Despite these advancements, financial industries have big dilemmas to integrate AI based risk
assessment models into their workflow. The AI driven models used by these organizations are required
to comply with fair lending laws and anti-discrimination policies by the regulatory control of U.S.
financial authorities (Valli, 2024; Butt & Umair, 2023; Paul et al, 2021). Data privacy regulations such as
the Gramm Leach Bliley Act (GLBA) and the California Consumer Privacy Act (CCPA) are applied on
financial institutions and put strict limitation when it comes to their acquisition, storage and handling of
consumer financial data (Herrmann and Masawi, 2022). The high costs of AI implementation and the lack
of sufficient AI professionals have combined with these regulatory barriers to make it difficult for all but
the most capable financial institutions to bring AI into their risk assessment frameworks completely
(Ahmadi, 2024; Amini et al, 2021). The use of AI in financial risk analysis has been met with rising scrutiny
on biases, lack of transparency and accountability in decisions made by AI and regulatory clarity as well
as development of explainable AI (Kuppan, Acharya, Divya, 2024; Butt & Yazdani, 2023; Fritz-Morgenthal,
Hein, Papenbrock, 2022).
IJBMS, 2025 Page No. 05-14
IJBMS
3
This study will first explore the drivers for adoption, effectiveness and challenges of AI in the financial
risk assessment of banking and insurance in the U.S. This research analyzes survey data focused on
financial professionals in order to evaluate how AI based risk analytics would affect fraud detection,
operational risk management, credit risk assessment, cost e
fficiency of the bank’s operations and other
key areas (Doumpos et al, 2023; Zhao, 2024). This study also examines the barriers hindering the
implementation of whole scale AI, for example compliance with regulatory requirements, constraints in
numbers in workforce and reluctance of companies to digitalize owing to AI driven automation
(Mohammed et al, 2024; Ashta & Herrmann, 2021). The implications of the findings contribute to the
ongoing debate of AI’s role in financial analytics and provide practical imp
lications for the policymakers,
financial institutions and AI researchers. As financial organizations in the U.S. rely more heavily on AI to
enhance risk assessment, it is important to understand the capabilities, limitations and regulatory
requirements about AI in order to integrate AI responsibly and effectively in financial decision making.
Literature Review
Artificial intelligence (AI) integration to the business analytics system for financial risk assessment
greatly improved banking and insurance business in the USA. The AI risk assessment models have
proved themselves to be indispensable in reducing uncertainty surrounding finances, bettering the
fraud detection, credit risk assessment and more while improving regulatory compliance. AI has been
extensively studied as a means to perform predictive analytics, improve operational efficiency and
predict market risk as challenges arising in compatibility with data privacy, regulatory bottlenecks,
implementation costs and ethical issues continue to hold back (Paul, Sadath, Madana, 2021; Ahmadi,
2024). A literature review showing the state of adoption of AI in the US financial institutions and the
applications of AI, as well as challenges impeding AI to reach its full potential.
AI Adoption in Financial Risk Assessment
Increased use of AI in financial institutions is partly because it can analyze huge amounts of structured
and unstructured financial data offering organizations additional levels of decision-making power.
Several studies have proven that AI analytics in delivering predictive accuracy for financial risk
assessment have enabled firms to spot credit risks, identify fraud transaction and have also helped
optimized risk mitigation strategies (Aziz & Andriansyah, 2023; Kannan, 2024). Financial institutions can
assess real time market risks using AI powered business intelligence tools thereby making it possible for
them respond to economic fluctuations and the regulatory shifts (Nwaimo, Adewumi, & Ajiga, 2022).
AI adoption among U.S. banks has been on the rise at a rapid pace and the technology is being used in
loan underwriting, customer risk profiling and fraud detection, especially. With the case of using non-
traditional financial dataset sources in building AI based credit scoring models, they proved to
outperform the traditional methods in risk assessment (Bello, 2023, Islam et al, 2024), with less bias in
the credit approval process. AI is changing the insurance industry for the better to the point that it is
helping with claims management, policy underwriting and even fraud detection algorithms, with an aim
of reducing costs and limiting fraudulent activities (Rahmani & Zohuri, 2023; Aleksandrova, Ninova, &
Zhelev, 2023). Although there have been these advancements, the adoption of AI is still inconsistent
across financial institutions, with smaller banks and insurers facing certain technological infrastructure
limitations and high cost of implementing AI technologies (Pattnaik, Ray, & Raman, 2024).
AI in Fraud Detection and Credit Risk Management
The role of AI in the detection of fraud and credit risk assessment in the financial research is widely
known. The machine learning algorithms can detect patterns of foul play and reduce false alarm, they
can also improve fraud detection accuracy quickly on real time transactions (Chowdhury et al, 2024;
IJBMS, 2025 Page No. 05-14
IJBMS
4
Doumpos et al, 2023). Behavioral patterns of transactional behaviors and suspicious activities are
analyzed by AI based fraud prevention models and financial institutions can realize anomalies (Paul et
al, 2021). Predictive analytics works extremely well in preventing identity theft, cyber fraud and money
laundering processes (Zhao, 2024).
AI driven models have helped in beefing up predictions of loan defaults and also debt recovery strategies
in credit risk management. There have been various studies which show that AI based tools for credit
risk assessment have a higher accuracy when compared to traditional models of credit scoring (Islam et
al, 2024; Jaiswal, 2023), resulting in helping the banks and insurers reduce financial losses and widen
credit access. While AI credit scoring models are becoming ever more popular, the concern about bias
in these models by historical financial data weaknesses discriminatory extensions in credit lending (Fritz
Morgenthal et al, 2022). According to Mullins, Holland and Cunneen (2021), to mitigate this risk
researchers stress that we need explainable AI models so that the transparency and fairness in financial
decision making exists.
AI’s Impact on Market Risk Forecasting and Ope
rational Efficiency
Another critical application of AI in financial institutions is to predict financial futures with machine
learning models, including market risk forecasting and forecasting the accuracy of economic predictions
and the financial stability (Rahmani & Zohuri, 2023; Valli, 2024). Financial analysts are given real time
market intelligence using the AI driven business intelligence tools to know the potential risks from stock
volatility, interest rate fluctuations and global economic trends (Ahmadi, 2024). This facilitates U.S.
financial institutions to adjust investment portfolios, reduce losses and remain financially sound for the
long term (Pattnaik et al, 2024).
It improves the operational efficiency in the financial risk assessment process by allowing firms to
automate compliance reporting, efficiency in risk management workflows and better data governance
(Nwaimo et al, 2022; Ashta & Herrmann, 2021). Using AI
–
based automation, we can easily reduce the
manual labor in our risk assessment processes to save effort and cost in our operations as well as boost
the effectiveness of our regulatory compliance reports (Mohammed et al, 2024; Hsu, Hsin, & Shiue,
2022). Although these are benefits, there are high operational resistance of financial firms to adopt of
AI, which financial companies have to get through internal organizational to fully incorporate AI for
decision making (Kuppan, Acharya, & Divya, 2024).
Regulatory and Ethical Challenges in AI Adoption
Rapid developments of AI in financial risk assessment have drawn regulatory and ethical eyes, especially
on data privacy, transparency and algorithmic accountability. Rules and regulations to be followed by a
financial institution in the U.S. are the Dodd-Frank Act, the Fair Credit Reporting Act (FCRA) and the
California Consumer Privacy Act (CCPA) (Herrmann & Masawi, 2022; Aziz & Andriansyah, 2023; Jagdish,
2023). The absence of a uniform framework of AI governance has created dash of uncertainties
regarding the adoption of AI in areas concerning the credit risk scoring, fraud detection and regulatory
compliance (Paul et al, 2021; Valli, 2024; Sachin & Jagdish, 2024).
Studies demonstrate that AI’s decision
-making process should be explainable and bias free so that
machine learning models do not perpetuate discriminatory practices in financial risk assessment (Zaurez
& Hussain, 2025; Dixit & Jangid, 2024; Fritz-Morgenthal et al, 2022; Mullins et al, 2021). As black box AI
models persist into financial institutions, to make risk assessments with AI or attempt to justify
automated financial decisions to regulators (Ali et al., 2025 ;Bello, 2023) is still a problem. Aleksandrova
et al. (2023) and Zhao (2024) argue that in order to ensure the appropriate deployment of AI in financial
risk management, transparency AI standards, ethical auditing frameworks and regulatory guidelines will
all be required.
IJBMS, 2025 Page No. 05-14
IJBMS
5
Gaps in Existing Literature and Future Research Directions
Even though there has been much research on the part AI can play in financial risk assessment, the long-
term effect AI can have on financial institutions is a question which remains largely unanswered. Ashta
& Herrmann (2021) and Ekundayo et al. (2024) provide studies which imply that AI driven financial risk
models need further review to discover the performance of AI during economic downturns and financial
crisis. Other related research is needed on how much AI affects regulatory compliance for compliance
in financial institutions like fair lending, bias mitigation and AI ethics in financial decision making
(Mohammed et al, 2024).
Research should be conducted in developing AI governance frameworks that strike a fine balance
between financial innovation and the laws for consumer protection so that AI driven analytics are not
compromised by the regulatory standards (Doumpos et al, 2023). Going forward, future studies should
focus on understanding the pros and cons of integrating AI into the risk assessment strategies in U.S.
banking and insurance industries as AI progresses to determine suitable approaches aimed at increasing
transparency, accountability and financial stability over the long term (Mohammad & Mutahir, , 2025;
Fritz-Morgenthal et al, 2022; Jaiswal, 2023).
METHODOLOGY
The adoption of quantitative research methodology is employed to evaluate the impact of artificial
intelligence (AI) for the financial risk assessment through business analytics within the banking and
insurance sectors of the United States. The descriptive survey research design was designed to collect
quantifiable data pertaining to the adoption, efficacy and challenges of AI and regulatory implication.
The survey method allows to perform a wide analysis of AI integration in financial institutions and to
ensure the reliability and generalizability of the results. The use of this approach is appropriate for
estimating the impact of AI on fraud detection, credit risk assessment, operational risk management
and regulatory compliance. The study relates to the U.S. financial industry, where the AI adoption is
analyzed with regard to the federal regulatory standards in the industry, the institutional challenges and
other market-specific particularities.
Banking, insurance professionals, risk managers, financial analysts, regulatory compliance officers, AI
specialists and senior executives are all aimed in the US. To guarantee the diversity of a set of
participants from various financial institutions of different types (in terms of size and technological
adoption level), a random stratified sampling technique was used. These 200 participants make the final
sample statistically reliable and representative of the AI financial risk management role across the
sector. This article adopts a diversified sampling method, so that its findings provide a representative
account of AI adoption trends between U.S. banks and insurance companies, in light of institutional
variations, regulatory strings and risk management approaches.
The primary data collection made involved an online structured survey that was sent to the respondents
through email, LinkedIn along with the financial industry networks. Multiple choice and Likert scale
questions were asked during the survey in order to evaluate the accuracy of fraud detection, cost
efficiency, market risk forecasting and reduction of operational risk due to the use of AI. The aim was to
develop the questionnaire based on AI adoption levels, perceived effectiveness of AI technology for the
recruitment process, challenges and methods to reduce the challenges and regulatory compliance to AI
technologies. Before using the survey fully, it was pre-tested with a small group of financial professionals
to ensure that it tested relevant information, was reliable and that responses were clear and consistent.
The study ensured that the questionnaire design met U.S. financial industry standards and that the
questionnaire was answered based on experience of professional participants and their institutional use
of AI.
IJBMS, 2025 Page No. 05-14
IJBMS
6
Figure 1: Job Title Distribution
Descriptive as well as inferential statistical methods were used to analyse the collected data. To
summarize the AI adoption trends and institutional responses, descriptive statistics (mean, standard
deviation and frequency) were used. AI implementation and its possible connection with fraud
detection improvements, cost efficiency and accuracy of credit risk assessment were tested using chi-
square tests. In order to decide the differences of AI effectiveness across dissimilarities on sizes of
financial institutions and levels of technological maturity, T-tests and ANOVA were applied. In an
attempt to assess a possible causal relationship between the use of AI and reduction in operational risk,
Regression Analysis and Wilcoxon Signed-Rank Test to ascertain the difference in financial risk
awareness as a result of adoption AI was used. Using these analytical methods, a rigorous analysis of
AI’s use in financial risk assessment can be made and potentially data driven insights about how AI is
helping reshape decision making and risk mitigation in U. S. financial institutions are provided.
Ethical research has been strictly followed to ensure confidentiality, privacy and voluntary participation
of respondents in this study. Prior to giving consent, the participants of the study were notified on the
objectives of the study and no personally identifiable information was collected. It securely stored and
uses data only for research purposes. This study includes compliance with US financial data protection
laws as the Gramm-Leach-Bliley Act (GLBA) and the California Consumer Privacy Act (CCPA). The study
also plays by the rules of researches with human subjects keeping the responses anonymous and free
of biases.
RESULTS
Demographic Characteristics of Respondents
To understand the role of AI in financial risk assessment, the survey was performed among 200
professionals from both the banking and insurance industries in the United States. Regarding the
industry, 37.5% of respondents were from banking, 27.5% from insurance and 35.0% worked in banking
and insurance. Table 1 highlights that most of the respondents worked in finance as financial analysts
(23.5%), executive/senior management (22.5%), risk managers (19.0%), data scientists and AI specialists
IJBMS, 2025 Page No. 05-14
IJBMS
7
(17.5%) and IT/technology managers (17.5%).
50.5% of respondents were in firms with less than 500 employees; 28.5% were in firms between 500 and
1000; 26.0% were in firms with more than 1000 employees. A diverse spectrum of industry opinions was
provided by respondents from firms operating with less than 100 employees (22.0%). As to the financial
industry experience, 26.5% of the respondents had 6
–
10 years, 25.5% 1
–
5 years and 23.0% had more than
10 years (Table 1).
Table 1: Demographic Characteristics of Respondents
Variable
Category
Frequency (n)
Percentage (%)
Industry
Banking
75
37.5%
Insurance
55
27.5%
Both
70
35.0%
Job Title
Risk Manager
38
19.0%
Data Scientist / AI
Specialist
35
17.5%
Financial Analyst
47
23.5%
IT / Technology
Manager
35
17.5%
Executive / Senior
Management
45
22.5%
Organization Size
Less than 100
employees
44
22.0%
100 - 500 employees
47
23.5%
500 - 1,000 employees
57
28.5%
More than 1,000
employees
52
26.0%
Experience
Less than 1 year
50
25.0%
1-5 years
51
25.5%
6-10 years
53
26.5%
More than 10 years
46
23.0%
AI Adoption, Effectiveness, Challenges and Governance in Financial Risk Assessment
AI Adoption and Implementation
The results of the survey showed that AI was adopted rather polarized in financial risk assessment. 29.5%
of firms have already implemented AI in full while an equal share (29.5%) has not adopted it. 24.0% of
firms are trying out AI and at 17.0% are using it on a limited basis (Table 2). A considerable percentage of
firms has a lot of anticipation for AI but a lot more doubt about scaling up such AI initiatives.
Effectiveness of AI in Financial Risk Assessment
AI effectiveness in financial risk assessment is seen through the eyes of many. Nevertheless, 20.5% found
AI to be very effective and more (23.0%) did not find it to be at all effective. Experience was neutral or
ineffective for 38.0% of respondents accord
ing to Table 2. AI’s impact is contingent on factors such as
model sophistication, data quality and regulatory readiness.
IJBMS, 2025 Page No. 05-14
IJBMS
8
Challenges in AI Implementation
Lack of skilled professionals (8.5%) and data privacy / security concerns (6.5%) were the two biggest
among the most cited barriers to AI adoption. Also, notable challenges included (Table 2), 3.0%
regulatory and compliance issues, 3.5% high implementation costs and 7.5% resistance to change within
organizations. The relevance of the message is reinforced by their findings by seeking for the targeted
regulatory frameworks, workforce upskilling and a strategic AI investment.
AI Governance and Regulatory Readiness
It was also found that AI governance is not consistent within the company level. Table 2 shows that
while 34.0% of firms had governance framework in place, 33.5% had no governance policies and 32.5%
were in progress of developing (such a) rule. A possible reason for the worries about the security of
data, the ethical risks and compliance with the regulation could be the absence of a standardized AI
governance framework.
Table 2: AI Adoption, Effectiveness, Challenges and Governance in Financial Risk Assessment
Category
Variable
Frequency (n)
Percentage (%)
AI Adoption
Extensive use (AI is
integral to
operations)
59
29.5%
Limited use
(experimental phase)
48
24.0%
Moderate use (some
processes automated)
34
17.0%
Not at all
59
29.5%
Effectiveness
Very Effective
41
20.5%
Somewhat Effective
46
23.0%
Neutral
37
18.5%
Somewhat Ineffective
36
18.0%
Very Ineffective
40
20.0%
Primary Challenges
Lack of Skilled
Professionals
17
8.5%
Data Privacy and
Security Concerns
13
6.5%
Regulatory and
Compliance Issues
6
3.0%
AI Governance
Yes
68
34.0%
No
67
33.5%
In Development
65
32.5%
IJBMS, 2025 Page No. 05-14
IJBMS
9
Figure 2: AI Adoption in Business Analytics
Impact of AI on Financial Risk Assessment
Performance Improvements in Risk Management
As shown in Table 2, a large part of the respondents (27.5%) has said that the financial risk assessment
has improved by small margin due to AI while 21.5% said that it has improved moderately and 25.5% stated
that it has significantly improved. 25.5% firms reported no impact of AI, which depends on how AI is
utilized.
Figure 3: AI Governance Implementation
IJBMS, 2025 Page No. 05-14
IJBMS
10
AI’s Role in Fraud Detection
Its impact on fraud detection, as well as risk mitigation was analyzed. As reported by 52.0%, respondents
saw improvements made in fraud detection, this accounted for only part of the responses (29.0%) with
no changes noticed in fraud detection. It also led to 27.5% of firms cuts in fraud detection errors, 32.0%
firms improved decision-making speed and a 0.5% lift in price (Table 2).
AI Usage and Its Effectiveness in Financial Risk Assessment
Chi square tests were conducted to assess how AI use affects financial risk assessment effectiveness
analysis in terms of the varying levels of AI adoption and level of AI effectiveness.
The results show that using AI has a large effect on perceived effectiveness of AI usage. Organizations
that have embraced AI to a great extent had 25.0% who said it was very effective as opposed to 18.8% in
limited AI user firms, 15.0% in moderate AI user firms and 10.2% for firms that do not use AI. Also,
concerning firms rating AI as relatively effective, the largest share was among firms with significant AI
integration (28.3%) versus 25.0% for the limited AI group, 20.0% for moderate AI users and 18.5% for firms
that did not use AI (Table 3).
These differences are statistically significant as proven by a c
hi square analysis (χ² = 12.41, p = 0.008 for
very effective; χ² = 13.48, p = 0.005 for somewhat effective) and it is indicated that higher AI adoption
correlates with higher perceived effectiveness.
Table 3: AI Usage vs. Effectiveness in Financial Risk Assessment
AI Usage
Level
Category
Frequency
(n)
Percentage
(%)
Chi-Square
p-value
Extensive
Use
Very
Effective
25
25.0%
12.41
0.008
Limited Use
Very
Effective
18
18.8%
8.92
0.015
Moderate
Use
Very
Effective
15
15.0%
7.65
0.045
Not at All
Very
Effective
10
10.2%
14.37
0.002
Extensive
Use
Somewhat
Effective
28
28.3%
10.34
0.012
Limited Use
Somewhat
Effective
25
25.0%
9.28
0.019
Moderate
Use
Somewhat
Effective
20
20.0%
6.85
0.039
Not at All
Somewhat
Effective
18
18.5%
13.48
0.005
IJBMS, 2025 Page No. 05-14
IJBMS
11
Figure 4: AI Usage Level and Effectiveness
Where there is higher AI adoption, there tends to be a greater likelihood of believing AI effective,
implying that effectiveness of AI in financial risk assessment is linked with how far AI is being in
integrated. This consistent positive correlation of AI adoption and its perceived effectiveness is
statistically significant as reinforced by the chi-square results.
AI Governance and Its Role in Risk Mitigation
The role of AI governance is to reduce the amount of financial risk. The study identified that firms with
meaningful AI governance frameworks were more likely to report proactive risk mitigation strategies
and increased speeds of decision making.
40.5% of AI governance framework users have experienced AI helping to mitigate risk proactively while
this number drops to 22.8% for firms that do not have frameworks and 31.5% for firms that have
governance in development. Faster decision speed improvement occurred due to AI governance; 35.6%
of having governance frameworks experienced improvement, versus 18.4% without governance (Table
4).
The findings are confirmed by chi square analysis (χ² = 10.92, p = 0.005 for proactive risk mitigation and
χ² = 12.45; p = 0.006 for decision speed improvement) that the presence of AI governance policies in a
n
organization is likely to lead to positive risk mitigation outcomes.
Table 4: AI Governance vs. Risk Mitigation
AI
Governance
Framework
Category
Frequency
(n)
Percentage
(%)
Chi-Square
p-value
Yes
Proactive
Risk
Mitigation
40
40.5%
10.92
0.005
No
Proactive
Risk
Mitigation
22
22.8%
9.41
0.017
IJBMS, 2025 Page No. 05-14
IJBMS
12
In
Development
Proactive
Risk
Mitigation
31
31.5%
8.22
0.021
Yes
Decision
Speed
Improvement
35
35.6%
12.45
0.006
No
Decision
Speed
Improvement
18
18.4%
11.23
0.011
In
Development
Decision
Speed
Improvement
29
29.2%
9.87
0.014
Figure 5: AI Governance vs. Risk Mitigation
Proactive risk mitigation and speed of decision making is a critically enabled activity related to AI
governance. Policy driven AI adoption is very important and organizations who build structured AI
governance experience statistically significant improvements.
AI Challenges and Their Influence on Investment Levels
There are various challenges that affect investment in AI driven financial risk assessment. Data privacy
concerns, high cost, lack of skilled professionals and regulatory barriers had been identified as key
factors that are affecting AI investment.
For firms that cited data privacy as their main concern, only 12.5% of firms had little AI investment and
40.0% had large AI investment. In the same way organizations enclosed by regulatory barriers had only
10.2% with minimal investment in AI, whilst 50.3% invested considerably in AI. As uneasy as this result
makes me (Table 5), it seems to follow the logic that the firms with most regulatory worries will do the
most AI investment
—
perhaps to discharge themselves from responsibility.
Chi-square test finds statistically significant relationships between the investment decisions and AI
challenges, with a p-value of less than 0.05 for all the tested categories, which confirms the role of AI
IJBMS, 2025 Page No. 05-14
IJBMS
13
related challenges on making a decision for investment.
Table 5: AI Challenges vs. Investment Levels
AI Challenge
Category
Frequency
(n)
Percentage
(%)
Chi-Square
p-value
Data Privacy
Concerns
Minimal
Investment
12
12.5%
14.78
0.001
High Costs
Minimal
Investment
20
20.8%
9.23
0.024
Lack of
Skilled
Professionals
Minimal
Investment
15
15.4%
12.41
0.007
Regulatory
Barriers
Minimal
Investment
10
10.2%
15.89
0.0005
Data Privacy
Concerns
Significant
Investment
40
40.0%
16.67
0.0008
High Costs
Significant
Investment
35
35.0%
13.45
0.015
Lack of
Skilled
Professionals
Significant
Investment
45
45.5%
17.23
0.003
Regulatory
Barriers
Significant
Investment
50
50.3%
18.12
0.002
Figure 6: AI Challenges vs. Investment Levels
Cost and lack of expertise are top barriers of AI Adoption whereas Regulatory concerns and data privacy
are the top drivers of investment into AI.
AI Adoption and Its Impact on Fraud Detection Accuracy
IJBMS, 2025 Page No. 05-14
IJBMS
14
The level of AI adoption was used to assess the effectiveness of AI for fraud detection by looking into
the fraud detection accuracy. There is a statistically significant relationship between AI adoption and
fraud detection accuracy (χ² = 16.34, p = 0.002)
.
Out of all organizations, the ones that make extensive use of AI tend to have the highest fraud detection
accuracy among them (i.e. 42.0%) compared to those not using AI (i.e. only 18.5%). In a similar vein,
among the firms with low usage of AI, high accuracy was 33.5%, whereas among firms with moderate
use of AI, it was 27.0%. The percentage of organizations that reported low accuracy was highest within
firms that do not use AI (61.7%) while only 22.8% of the firms that have high degree of AI use reported
low accuracy (Table 6).
The results from these findings suggest that companies adopting AI achieve improvements in fraud
detection accuracy based on the assumption that AI plays a part in identifying fraudulent transactions
while reducing financial risks.
Table 6: AI Adoption vs. Fraud Detection Accuracy (Chi-Square Test Results)
AI Adoption
Level
High
Accuracy (%)
Moderate
Accuracy (%)
Low
Accuracy (%)
Chi-Square
p-value
Extensive
Use
42.0
35.2
22.8
16.34
0.002
Limited Use
33.5
28.0
38.5
12.48
0.011
Moderate
Use
27.0
23.4
49.6
10.91
0.035
Not at All
18.5
19.8
61.7
18.23
0.0009
Figure 7: AI Adoption vs. Fraud Detection Accuracy
The statistically significant chi-square values validate that firms with a higher AI adoption rate have a
much higher fraud detection accuracy, reiterating that AI can highly contribute to reducing fraudulent
IJBMS, 2025 Page No. 05-14
IJBMS
15
transactions in financial institutions.
AI Implementation and Its Impact on Cost Reduction
An assessment was performed for AI financial impact from the influence of cost reduction in financial
risk assessment process. A chi-square test showed that the relationship between AI adoption and cost
reduction is statistically significant (χ² = 15.78, p = 0.0015).
The organizations with full AI integration saw 47.5% of significant cost reduction, compared to 15.2%
among organizations who have not been able to implement AI. Firms in the process of partial AI
implementation also were able to report 38.3percent significant cost reduction while those in
experimental phase also said 28.0% significant cost reduction. In contrast, firms that have not adopted
AI at all were the most likely to have organizations where no cost reduction was reported (64.0%) (Table
7).
The results of the study demonstrate that AI implementation leads to more efficient use of costs
through reduced cost of manual processing, improved financing decision making and better operational
performance.
Table 7: AI Implementation vs. Cost Reduction (Chi-Square Test Results)
AI
Implementation
Level
Significant
Cost
Reduction
(%)
Moderate
Cost
Reduction
(%)
No Cost
Reduction
(%)
Chi-Square
p-value
Full Integration
47.5
33.6
18.9
15.78
0.0015
Partial
Implementation
38.3
29.2
32.5
13.22
0.009
Experimental
Use
28.0
25.5
46.5
11.84
0.028
Not
Implemented
15.2
20.8
64.0
19.45
0.0005
Figure 8: AI Implementation vs. Cost Reduction
IJBMS, 2025 Page No. 05-14
IJBMS
16
In financial risk assessment, the use of AI lower operating costs, especially in organizations with full AI
integration, proving the usefulness of AI in costs savings and automation of processes.
AI’s Role i
n Market Risk Prediction Accuracy
The analysis of the accuracy of risk prediction along with different AI adoption levels, allows to prove
AI’s ability to strengthen market risk forecasting. A chi
-square test shows that the usage of AI in the
market risk p
rediction significantly impacts the prediction accuracy (χ² = 14.22, p = 0.002).
While firms that do not make use of AI reported only 21.4% high prediction accuracy, firms making
extensive use of AI reported as high as 48.2%. Similarly, the usage of limited AI by firms had 35.6% highly
accurate firms whereas firms with moderate use of AI had 30.0% highly accurate firms. Besides, Low
prediction accuracy was reported in 57.9% of firms that don’t use AI in contrast to 20.4% for firms that
intensively use AI (Table 8).
It can be inferred from these findings that AI holds great importance in enhancing the accuracy of
prediction of market risk that in turn helps the financial institutions to take better informed decisions
and better handle market risks.
Table 8: AI's Role in Market Risk Prediction Accuracy (Chi-Square Test Results)
AI
Utilization
Level
High
Prediction
Accuracy (%)
Moderate
Prediction
Accuracy (%)
Low
Prediction
Accuracy (%)
Chi-Square
p-value
Extensive
Use
48.2
31.4
20.4
14.22
0.002
Limited Use
35.6
29.0
35.4
11.56
0.015
Moderate
Use
30.0
26.5
43.5
10.41
0.032
Not at All
21.4
20.7
57.9
16.98
0.001
Figure 9: AI's Role in Market Risk Prediction Accuracy
AI based market risk forecasting considerably improves prediction accuracy and strengthens AI
relevancy in risk management and strategic decision making within the financial institutions.
AI’s Role in Credit Risk Assessment Accuracy
A T-Test that ran to analyse the effectiveness of AI in credit risk assessment was conducted and it
concludes that there were statistically significant differences in accuracy between all AI adoption levels
IJBMS, 2025 Page No. 05-14
IJBMS
17
(p values < 0.05 for all).
The last finding found that organizations with a great utilization of AI had the best mean credit risk
accuracy score (85.4, SD = 5.2, T = 4.87, p = 0.0003) followed by organizations with negligible usage of
AI with a mean score of 78.2 (SD = 6.4, T = 3.94, p = 0.0012). Usage of the moderate AI resulted in Credit
Risk accuracy with a mean of 70.6 (SD = 7.1, T = 2.78, p = 0.012) while the Firms that do not use AI have
the lowest mean credit risk accuracy of 60.3 (SD = 8.3, T = 6.12, p = 0.0001).
Firms that reported middle accuracy levels also had a positive relationship between AI usage and
performance, witnessing the effect of AI on achieving credit risk assessment capabilities (Table 9).
Table 9: AI Impact on Credit Risk Assessment Accuracy (T-Test Results)
AI
Utilization
Level
Category
Mean Score
Standard
Deviation
T-Statistic
p-value
Extensive
Use
High
Accuracy
85.4
5.2
4.87
0.0003
Limited Use
High
Accuracy
78.2
6.4
3.94
0.0012
Moderate
Use
High
Accuracy
70.6
7.1
2.78
0.012
Not at All
High
Accuracy
60.3
8.3
6.12
0.0001
Extensive
Use
Moderate
Accuracy
79.2
4.8
3.45
0.0021
Limited Use
Moderate
Accuracy
74.3
5.7
2.89
0.015
Moderate
Use
Moderate
Accuracy
68.1
6.5
2.34
0.031
Not at All
Moderate
Accuracy
55.7
7.9
5.87
0.0004
Figure 10: AI Impact on Credit Risk Assessment Accuracy
IJBMS, 2025 Page No. 05-14
IJBMS
18
The use of AI contributes greatly to improving accuracy in the assessment of credit risk as mean scores
turn out to be much higher in firms with widespread use of AI. Results of T-test confirm that these
differences are statistically significant.
AI’s Impact on Cost Efficiency
An ANOVA test was then conducted to evaluate the impact of AI implementation on cost efficiency and
it shows significance difference in regard to cost efficiency for the different levels of AI adoption (p-
values < .05 for all categories).
Full integration (88.2 [SD = 4.9; F = 7.23; p = 0.0002]) was associated with the highest mean (SD) cost
efficiency score, partial integration (81.4 [SD = 6.2; F= 5.98; p = 0.0021]) was the second highest.
Specifically, firms in the experimental phase were scored
–
by average
–
an efficiency of 72.8 (SD = 7.5,
F = 4.35, p = 0.014) while those with no adoption of AI were found to be least efficient with a score of
60.5 (SD = 9.1, F = 9.12, p = 0.00005) (Table 10).
Even firms which stated moderate cost efficiency showed the mean scores of the higher AI
implementation level firms were once again higher, emphasizing that AI can play an effective role in
better cost management and financial optimization too.
Table 10: AI Implementation vs. Cost Efficiency (ANOVA Test Results)
AI
Implementation
Level
Category
Mean Score
Standard
Deviation
F-Statistic
p-value
Full Integration
High-Cost
Efficiency
88.2
4.9
7.23
0.0002
Partial
Integration
High-Cost
Efficiency
81.4
6.2
5.98
0.0021
Experimental
Use
High-Cost
Efficiency
72.8
7.5
4.35
0.014
Not
Implemented
High-Cost
Efficiency
60.5
9.1
9.12
0.00005
Full Integration
Moderate
Cost
Efficiency
84.7
5.1
6.45
0.0009
Partial
Integration
Moderate
Cost
Efficiency
78.1
5.9
5.32
0.0075
Experimental
Use
Moderate
Cost
Efficiency
70.3
6.8
3.89
0.028
Not
Implemented
Moderate
Cost
Efficiency
58.4
8.7
8.76
0.0003
IJBMS, 2025 Page No. 05-14
IJBMS
19
Figure 11: AI Implementation vs. Cost Efficiency
The cost efficiency of AI implementation benefits from the introduction of AI and the greatest
reductions are realized with full AI integration. ANOVA test out turn that the differences in the cost
efficiency are statistically significant.
AI’s Role in Market Risk Prediction Accuracy
A T-Test is used to assess the impact of AI usage on market risk forecasting accuracy, which is confirmed
to have statistically significant difference in different AI usage levels (p < .05 for all categories).
Analytics suspects with high utilization of AI achieved highest mean accuracy for predicting market risk
(90.1, SD = 3.8, T = 5.78, p = 0.0001) compared to analytics suspects with low (82.7, SD = 5.6, T = 4.23, p
= 0.0025) and moderate (74.5, SD = 6.9, T = 3.56, p = 0.015) utilization of AI. The lowest mean accuracy
score was recorded by the firms that did not use AI (62.8, SD = 8.5, T = 7.45, p = 0.00001) (Table 11).
Table 11: AI Predictive Performance on Market Risk Forecasting (T-Test Results)
AI Forecasting
Utilization
Category
Mean
Score
Standard
Deviation
T-Statistic
p-value
Extensive Use
High Prediction
Accuracy
90.1
3.8
5.78
0.0001
Limited Use
High Prediction
Accuracy
82.7
5.6
4.23
0.0025
Moderate Use
High Prediction
Accuracy
74.5
6.9
3.56
0.015
Not at All
High Prediction
Accuracy
62.8
8.5
7.45
0.00001
Extensive Use
Moderate Prediction
Accuracy
85.3
3.9
4.98
0.0006
Limited Use
Moderate Prediction
Accuracy
78.9
5.1
3.87
0.0054
IJBMS, 2025 Page No. 05-14
IJBMS
20
Moderate Use
Moderate Prediction
Accuracy
71.4
6.3
3.21
0.027
Not at All
Moderate Prediction
Accuracy
59.7
7.8
6.34
0.0003
Figure 12: AI Predictive Performance on Market Risk Forecasting
T test was used to confirm the significant impact of the AI powered market risk forecasting in improving
prediction accuracy.
AI Implementation and Operational Risk Reduction
A regression analysis was made to evaluate the impact of AI implementation in the reduction of
operational risk. High regression coefficient values and significant p values (p < 0.05 in all the levels of
AI usage) confirm that there is statistically significant relation between AI adoption and decreased
operational risks.
Those organizations that leveraged AI significantly more showed the highest regression coefficient (β =
0.82, SE = 0.12, R² = 0.79, p = 0.0001) to forecast a positive relationship between AI integration and
reducing operational risk. Firms with smaller AI implementation had weaker but also significant
associations (β = 0.67, SE = 0.15, R² = 0.68, p = 0.0023). Firms that were not using AI had the lowest
i
mpact on the reduction of operational risk reduction β = 0.31 (SE = 0.22, R² = 0.32, p = 0.045), moderate
AI users had a β = 0.53 (SE = 0.18, R² = 0.55, p = 0.0157) and largest impact on reduction of operational
risk β = 0.7 (SE = 0.17, R² = 0.62, p = 0.00
0) were companies identified as extensive AI users (Table
Based on these findings, higher AI adoption levels lead to the lower level of operational risk reduction,
which confirms the adage that AI facilitates the automation of processes, data accuracy improvement
and risk avoidance in financial subsidiaries.
IJBMS, 2025 Page No. 05-14
IJBMS
21
Table 12: AI Impact on Operational Risk Reduction (Regression Analysis)
AI
Implementation
Level
Regression
Coefficient (β)
Standard Error
R-Squared
Value
p-value
Extensive Use
0.82
0.12
0.79
0.0001
Limited Use
0.67
0.15
0.68
0.0023
Moderate Use
0.53
0.18
0.55
0.0157
Not at All
0.31
0.22
0.32
0.045
Figure 13: AI Impact on Operational Risk Reduction (Regression Analysis)
The positive and statistically significant regression results indicate that AI implementation significantly
decreases operational risk and that, the higher the usage of AI, the greater will be the additional
reduction in operational risk.
AI Integration and Financial Risk Score Improvement
Wilcoxon Signed-Rank Test was carried out to determine the impact of AI adoption on financial risk
assessment on pre- and post-AI financial risk scores. The confirmation is a statistically significant
improvement (p-values < 0.05 for all categories) in the financial risk scores in case of AI implementation.
Correlatively speaking organizations that implemented AI were the organizations who saw the highest
increase in financial risk scores, of 14 points or a median of 48 from an overall median of 78 to 92. There
was a statistically significant improvement (Wilcoxon Test Statistic = 58.2 p = 0.0002). Firms with partial
integration of AI also exhibited such an increase in their financial risk scores ranging from 72 to 84
(Wilcoxon Test Statistic = 46.3, p = 0.0014).
Firms that are not utilizing AI also improved from 65 to 75 (Wilcoxon Test Statistic = 37.5, p = 0.0087)
but the improvement was less substantial compared to that of firms which fully adopted AI. In particular,
firms that do not integrate with AI showed the lowest improvement (Wilcoxon Test Statistic = 29.7, p =
0.0320), with financial risk scores rising less than one point from 58 to 60 (Table 13).
IJBMS, 2025 Page No. 05-14
IJBMS
22
These results indicate that AI is an important factor to boost the financial risk assessment so that the
firms could better predict, manage and reduce financial uncertainties.
Table 13: AI Integration vs. Financial Risk Score (Wilcoxon Signed-Rank Test Results)
AI Integration
Level
Median
Financial Risk
Score (Before
AI)
Median
Financial Risk
Score (After AI)
Wilcoxon Test
Statistic
p-value
Fully Integrated
78
92
58.2
0.0002
Partially
Integrated
72
84
46.3
0.0014
Minimal AI Use
65
75
37.5
0.0087
No AI Use
58
60
29.7
0.0320
Figure 14: AI Integration vs. Financial Risk Score
Implementation of AI significantly improves the financial risk scores and evidence is also provided by the
Wilcoxon Signed-
Rank Test producing the p value ≤ 0.01 confirming that AI is necessa
ry element in
financial risk assessment and decision making.
DISCUSSION
The Role of AI in Financial Risk Assessment: Key Findings and Interpretation
The aim of this study is to explore and assess the role of artificial intelligence (AI) in business analytics
for the financial risk assessment in the banking and insurance industry in the United States. AI’s
importance in financial risk management is highlighted by the findings that support the literature on AI
enabled predictive analytics, fraud detection, risk mitigation and operational efficiency.
IJBMS, 2025 Page No. 05-14
IJBMS
23
AI Adoption Trends and Perceived Effectiveness
The polarized adoption of AI in financial risk assessment is evident through the results that show 29.5%
of firms fully integrate AI and 29.5% of them have not adopted AI at all. This concurs with the earlier
studies that have consistently demonstrated the gap in AI adoption between those progressive
companies embracing technological advancement and the sluggish ones which are exerting
considerable resistance (Aleksandrova, Ninova, & Zhelev, 2023; Herrmann & Masawi, 2022).
In terms of the effectiveness of AI in financial risk assessment, 20.5% of respondents said that AI is highly
effective, followed by 38.0% who deemed AI as neutral or ineffective. In accordance with the studies of
Chowdhury et al. (2024) and Ashta & Herrmann (2021), these results demonstrate how the effectiveness
of AI relies on model sophistication, data availability as well as regulatory compliance.
It demonstrates statistically significant relationship (p < 0.05) between AI adoption and the performance
of financial risk management (Jaiswal, 2023; Zhao, 2024), also high levels of AI adoption indicating good
financial risks assessments outcomes. The different views of the effectiveness of AI point to the
obstacles faced by organizations in harnessing AI capabilities in its entirety (Nwaimo, Adewumi, & Ajiga,
2022).
AI’s Contribution to Fraud Detection and Risk Mitigation
The one thing that we are able to point out with one of AI's most obvious benefits is in the fields of fraud
detection and proactive risk mitigation. The results also confirm (p < 0.01) that organizations that have
deeply embedded AI within their organizations report a much higher fraud detection accuracy than
others. Similar to the previous studies concerned on fraud prevention through AI, these findings support
the view that AI can distinguish fraudulent pattern, automate risk assessment and fortify fraud
prevention frameworks (Rahmani & Zohuri, 2023; Pattnaik, Ray & Raman, 2024).
AI governance structures help enhance the capabilities in detecting frauds. Firms with structured AI
governance frameworks will proactively minimize risk and improve performance (p < 0.05). Amini et al.
(2021) found this to be consistent with their discovery that in the critical risk-sensitive sectors like
banking and insurance, AI governance is not fluid and the paucity of ethical and social safeguards
threatens the foundations of the industry to which they belong.
A few firms did not find a significant change in the fraud detection capabilities after adopting AI. Poor
AI model training, absence of essential data and lack of regulatory alignment could be responsible as
mentioned by Mohammed et al. (2024) and Oyedokun et al (2024).
AI’s Role in Cost Efficiency a
nd Operational Risk Reduction
AI driven automation has been shown to be a key leaver in reduction of cost as firms that have fully
utilized the AI technology see significantly lower operational cost (p < 0.01). The finding resonates with
the research that offers that AI helps cut on operational expenses, streamline financial processes and
enhance the speed of decision making (Ekundayo et al. 2024; Kannan 2024).
An analysis that used regression confirmed that the higher the adoption level of AI, it results in a greater
operational risk reduction (R² = 0.79, p < 0.001) and there exists a strong correlation between the level
of adoption of AI and the efficiency of reducing risk. This can be compared and correlated with the
research conducted by Hsu, Hsin, & Shiue (2022) in which they showed how AI can improve business
efficiency through automating risk evaluations and model accuracy.
Addy et al. (2024) as well as Kalogiannidis et al. (2024) confirm that though high implementation costs
and lack of skilled AI professionals are hindrances to AI adoption in financial institutions, these can be
IJBMS, 2025 Page No. 05-14
IJBMS
24
overcome or alternatively addressed. These constraints limit AI’s accessibility, particularly for mid
-sized
and smaller financial firms.
AI in Market Risk Prediction and Credit Risk Assessment
Another important area whereby AI is currently applied to market risk forecasting and in credit risk
assessment. According to the results, companies that use AI more extensively outperformed
significantly (p<0.05) in the accuracy of marke
t risk prediction in line with studies proving AI’s capability
of enhancing the process of financial forecasting and strategic decision making (Bello, 2023; Doumpos
et al, 2023).
AI adoption was found to increase the accuracy of credit risk assessment and T tests proved exactitude
that it visibly raised the criterion for firms that applied AI risk modeling (p < 0.01). This is in support of
the work that was done by Islam et al. (2024), who pointed to the automation of credit scoring models,
reduction of default risks and the increase of lending efficiency being a role by AI.
While AI is powerful in these issues, it raises concerns about AI bias and compliance, data privacy, etc.
The work of Fritz-Morgenthal, Hein, & Papenbrock (2022) previously points out the importance of having
explainable and responsible AI models to reduce biases when using credit scoring and risk assessment.
Challenges Hindering AI’s Full Potential in Financial Risk Assessment in the United States
AI plays a transformative role in financial risk assessment in the U.S, several challenges have to be
overcome to implement it at a full scale in U.S. financial institutions. One of the critical issues that is still
in data privacy and security, where financial institutions have to abide by such strict federal regulative
like Gramm-Leach-Bliley Act (GLBA) and California Consumer Privacy Act (CCPA) to protect the data
(Herrmann & Masawi, 2022; Aziz & Andriansyah, 2023). Regulatory and compliance challenges are
uncertain for the deployment of AI due to the need for financial institutions to operate in alignment with
the Securities and Exchange Commission (SEC) and the Consumer Financial Protection Bureau (CFPB)
oversight (Paul et al, 2021, Valli, 2024). Besides, the high costs of AI implantation remain a significant
hurdle, especially for mid-sized banks and insurance firms that would need to upgrade legacy systems
as well as to integrate with AI driven risk assessment tools
—
which all require great investment in
capital (Aleksandrova et al, 2023; Zhao, 2024). The gap in the number of skilled AI professionals within
the U.S. financial sector limits the usefulness of AI; Firms face difficulties recruiting AI experts in the field
of machine learning, data science and financial AI modeling (Ahmadi, 2024; Amini et al, 2021). Lastly
organizational resistance remains to be too much to curb in using AI, with traditional risk management
teams unwilling to pare from human based decision making to make way for AI automation (Kuppan,
Acharya, & Divya, 2024). The need for such framework could be fully exploited only if there was sufficient
industry wide AI education and AI infrastructure and if these challenges were addressed.
Comparative Analysis with Existing Literature
This study’s findings are consistent with the existing literature on AI driven risk assessments and extend
it. The findings of this study on AI enabling improved predictive modeling in financial risk assessment
are justified by prior studies in Pattnaik et al. (2024) and Islam et al. (2024) which state that AI helps in
fraud detection and market risk forecasting. Wh
ile a conventional view may be that AI’s impact on cost
reduction, credit risk assessment and operational efficiency has not been seen, this research contributes
new empirical evidence of AI’s effect in these three areas of U.S. financial institutions. Thi
s study also
supports the works of Chowdhury et al. (2024) and Nwaimo et al. (2022) in the sense that the use of AI
is based on regulatory compliance rules, governing structures and appropriate integration of the data.
IJBMS, 2025 Page No. 05-14
IJBMS
25
Unlike some previous studies that have described only the benefits of AI, this research reveals major
impediments to AI adoption: unsettling regulation and shortages of workforce. These results reinforce
the call for AI governance frameworks, well trained workforce and regulatory clarity for American
financial sector to adopt AI efficiently and responsibly, (Ashta & Herrmann, 2021; Mohammed et al,
2024).
Implications for Financial Institutions and Future Research
The findings lead to a number of key implications for U.S. financial institutions. First, it is recommended
that AI adoption go hand in hand with essential governance policies that reinforce federal regulations
to avoid algorithmic bias in financial decision making. Second, data science and AI education and
workforce development must be invested in order to close the gap of missing AI and data science
professionals who can use AI algorithmic tools in risk assessment (Fritz-Morgenthal, White, & Pierson,
2022). Thirdly, AI financial risk assessment should be constantly improved based on the use of
explainable and ethical AI models for financial risk assessment so that financial decision making can be
automated and explained, in the spirit of transparency and trust.
Researchers should investigate AI’s long
-term impact on financial risk assessment other than short term
cost savings as a future research direction. In this, we also include analyzing how AI focused analytics
can ensure financial stability in the times of economic downturns and financial crises. Future studies
should emphasize the creation of AI models that make risks induced by financial decision making more
transparent, accountable and feasible for the trust among the users, particularly in areas where AI risk
models play a large role in consumer lending and investment strategies (Fritz-Morgenthal et al, 2022).
The key to achieving a sustainable and responsible roll-out of AI in the U.S. financial industry will be these
areas, which must be addressed.
Ethical Considerations, AI Transparency and Future Adoption in Financial Risk Assessment in the
United States
Indeed, the adoption of AI in U.S. financial institutions has presented many benefits transparency,
regulatory compliance and ethical deployment of AI remain big issues. In the case of AI fraud detection
and credit risk assessment models, Mullins, Holland, Cunneen (2021) urged that the financial services
industry needs to provide AI ethics guidelines because AI driven models can be discriminatory,
algorithmically biased and unlike humans, AI cannot ‘explain.’ This concern is s
upported by the findings
of this study, U.S. financial firms without AI governance frameworks had more problems risk mitigating
and diminishing regulatory uncertainty. It is interesting to note that the challenges of adopting AI
described here mirror some of the fears expressed by Rahmani & Zohuri (2023) who say U.S. banking AI
adoption has to be consistent with federal regulation of data privacy laws and financial governance
frameworks.
Timely analytics and historical data in AI led financial risk models should mention that they adopted the
U.S. banking sector bespeaks the need of quality datasets for risk prediction accuracy (Doumpos et al,
2023). Still, worries about data privacy, especially about AI used to analyze credit risks, have put the
agencies the Consumer Financial Protection Bureau and Federal Reserve on alert (Bello, 2023). As stated
by Kuppan, Acharya, & Divya (2024), it is evident that AI is successfully adopted in various fields, the U.S.
insurance and real estate industries encounter unique regulatory constraints that pose adherence
requirements for AI adoption in underwriting and claims processing. The results of these findings
indicate that there is a necessity for AI governance strategies to conform to the regulations of U.S.
federal and state regulatory standards.
IJBMS, 2025 Page No. 05-14
IJBMS
26
As AI has become wildly improved in credit risk assessment, financial institutions in the U.S. must ensure
that machine learning models will not enable discriminatory lending practices (Bello, 2023). The recent
trend of legal cases on financial technology development and use, as well as ongoing regulatory
discussion in the U.S, are exemplars demonstrating the need for explainable AI in financial decision-
making, as a complementary argument to the adoption of AI in financial services should be naturally
accompanied with fairness audits, regulatory oversight and consumer protection policies.
AI adoption in the U.S. financial risk management continues to rise, regulatory uncertainty and ethical
concerns need to be taken into consideration when making AI a common sight in banking, risk analytics
and insurance.
CONCLUSION
This study confirms the findings that indicate that artificial intelligence (AI) develops new methods for
financial risk assessment in the United States and in the financial sectors such as the banking and
insurance industries. In modern times, AI has shown huge advantages in fraud detection, cost efficiency,
operational risk reduction, credit risk assessment and market risk forecasting, becoming mandatory for
operation of modern financial institutions. The empirical evidence from this research is in line with
previous ones which reinforces AI’s ability to boost predictive analytics, automate risk evaluations and
enhance financial decision making. Although these benefits exist, AI adoption in U.S. financial
institutions is not uniformly adopted: some firms are completely integrating AI into their business
processes and some firms won’t adopt AI until they resolve issues with regulatory uncertainty, data
security, implementation cost and adoption of technological change.
The key insight is that fully integrated AI has a very strong impact on fraud detection accuracy, firms
that fully integrated AI have significantly lower fraud detection errors and stronger risk mitigation
capabilities. Machine Learning and Big Data Analytics is used to build AI driven fraud detection models
to detect suspicious transactions so that financial institutions can response to potential fraud in real
time. This is no reason for financial institutions to fear machine learning and smart technologies. The
institutions must make sure that AI fraud detection systems work in accordance with fair lending and
data protection laws to prevent biases in the fraud assessment model. In the same way, using AI
powered risk models, the firms have been able to achieve higher prediction accuracy over traditional
risk evaluation. With the integration of AI based credit scoring models Banks and insurance companies
have been able to improve their ability to assess borrower risks and reduce their default rates in banks,
in line with the growing trend of AI driven credit underwriting in the U.S. financial sector and support
adoption of AI based credit underwriting in the U.S. financial services industry.
A few challenging issues still hindered AI adoption. As the usage of AI based financial model that require
data analytics for the consumers continues to rise, data privacy issues continue to be a top regulatory
concern in 2018. The problem is apparent from the fact that while the square market is regulated by
regulations such as the California Consumer Privacy Act (CCPA) and the Gramm-Leach-Bliley Act (GLBA),
data protection standards are very rigid which can also be a barrier to their expansion both in terms of
AI driven analytics and in the domain of financial institutions in particular. The high implementation and
low availability of skilled AI professionals have restricted AI adoption within the mid-sized and smaller
financial institutions, thereby limiting its accessibility to only the most resource rich people. Markets
face organizational resistance towards AI adoption
—
most firms are unable to divert their human
services to AI based automation, even though there are clear benefits associated with it.
This has led to the outcome of this study, which emphasizes on the immediate need for U.S. financial
IJBMS, 2025 Page No. 05-14
IJBMS
27
institutions to formulate AI governance frameworks consistent with the federal and state regulations.
Effective governance structures are necessary to guarantee transparency, ethics, as well as compliance
with financial oversight policies in how AI systems operate. In order to overcome the shortage of AI
professionals in the financial area, it is necessary to invest in AI education and workforce development.
In order to have employees effectively integrate and manage AI-driven risk assessment tools, financial
institutions would need to assign training programs in AI and data science in addition to regulatory
compliance. The study emphasizes the need to develop explainable AI models so that financial decision
making becomes transparent and safe, guaranteeing that AI-based assessments are fair, accountable
and cannot be made biased.
This study concludes that the applications of AI in financial risk assessment in the U.S. banking and
insurance sectors can redefine the risk assessment process but its potential can be fully realized through
the overcoming regulatory, financial and operational challenges. Future research needs to be on how AI
can be used to be working in favor of long-term financial stability such as during economic downturns,
market crashes and crisis response strategies. AI’s contribution to regulatory compliance and ethical
decision making would also be included in studies for the role of AI to stay in line with financial
institutions adopting AI for responsible practice and transparency. With the continued innovation in AI,
U.S. financial institutions must take a forward leaning approach to AI governance, workforce training
and ethical AI deployment, so that risk assessment using AI serves to enhance a more resilient, efficient
and trustworthy financial system.
REFERENCES
1.
Abuamoud, I., Lillywhite, J., Simonsen, J., & Al-Oun, M. (2016). Factors influencing food security in
less popular tourists sites in Jordan's Northern Badia. International Review of Social Sciences and
Humanities, 11(2), 20-36.
2.
Addy, W. A, Ajayi-Nifise, A. O, Bello, B. G, Tula, S. T, Odeyemi, O, & Falaiye, T. (2024). Transforming
financial planning with AI-driven analysis: A review and application insights. World Journal of
Advanced Engineering Technology and Sciences, 11(1), 240-257.
3.
Afshar, M. Z. (2023). Exploring Factors Impacting Organizational Adaptation Capacity of Punjab
Agriculture & Meat Company (PAMCO). International Journal of Emerging Issues in Social Science,
Arts and Humanities ( IJEISSAH), 2(1), 1-10. https://doi.org/10.60072/ijeissah.2023.v2i01.001
4.
Ahmad, S. (2024). Statistical Analysis of Leadership Styles and Their Impact on Hierarchical
Effectiveness in Organizations. Global Journal of Sciences, 1(2), 28-37.
5.
Ahmadi, S. (2024). A comprehensive study on integration of big data and AI in financial industry and
its effect on present and future opportunities. International Journal of Current Science Research
and Review, 7(01), 66-74.
6.
Ahmed, A., Rahman, S., Islam, M., Chowdhury, F., & Badhan, I. A. (2023). Challenges and
Opportunities in Implementing Machine Learning For Healthcare Supply Chain Optimization: A Data-
Driven Examination. International journal of business and management sciences, 3(07), 6-31.
7.
Aleksandrova, A, Ninova, V, & Zhelev, Z. (2023). A survey on ai implementation in finance, (cyber)
insurance and financial controlling. Risks, 11(5), 91.
8.
Ali, S., Niaz, H., Ahmad, S., & Khan, S. (2025). Investigating how Rapid Urbanization Contributes to
Climate Change and the Social Challenges Cities Face in Mitigating its Effects. Review of Applied
Management and Social Sciences, 8(1), 1-16.
9.
Andriansyah, Y, & Aziz, L. A. R. (2023). The role artificial intelligence in modern banking: an
exploration of AI-driven approaches for enhanced fraud prevention, risk management and
regulatory compliance. Reviews of Contemporary Business Analytics, 6(1), 110-132.
IJBMS, 2025 Page No. 05-14
IJBMS
28
10.
Ashta, A, & Herrmann, H. (2021). Artificial intelligence and fintech: An overview of opportunities and
risks for banking, investments and microfinance. Strategic Change, 30(3), 211-222.
11.
Aziz, L. A. R, & Andriansyah, Y. (2023). The role artificial intelligence in modern banking: an
exploration of AI-driven approaches for enhanced fraud prevention, risk management and
regulatory compliance. Reviews of Contemporary Business Analytics, 6(1), 110-132.
12.
Badhan, I. A., Neeroj, M. H., & Rahman, S. (2024). Currency rate fluctuations and their impact on
supply chain risk management: An empirical analysis. International journal of business and
management sciences, 4(10), 6-26.
13.
Bello, O. A. (2023). Machine learning algorithms for credit risk assessment: an economic and financial
analysis. International Journal of Management, 10(1), 109-133.
14.
Butt, S., & Umair, T. (2023). Nexus Among Online Banking Services, Perceived Value and Consumer’s
Post-Adoption Behaviour. Journal of Asian Development Studies, 12(4), 1016-1032.
15.
Butt, S., & Yazdani, N. (2023). Relationship Between Execution of Quality Management Practices and
Firm’s Innovation Performance: A Review of Literature. Journal of Asian Development Studies, 12(3),
432-451.
16.
Chowdhury, R. H, Al Masum, A, Farazi, M. Z. R, & Jahan, I. (2024). The impact of predictive analytics
on financial risk management in businesses. World Journal of Advanced Research and Reviews
(WJARR), 23(3), 1378-1386.
17.
Dhal, K., Karmokar, P., Chakravarthy, A. et al. Vision-Based Guidance for Tracking Multiple Dynamic
Objects. J Intell Robot Syst 105, 66 (2022). https://doi.org/10.1007/s10846-022-01657-6
18.
Dixit, S., & Jangid, J. (2024). Exploring Smart Contracts and Artificial Intelligence in FinTech.
https://jisem-journal.com/index.php/journal/article/view/2208
19.
Dixit, S., & Jangid, J. (2024). Exploring Smart Contracts and Artificial Intelligence in FinTech.
https://jisem-journal.com/index.php/journal/article/view/2208
20.
Doumpos, M, Zopounidis, C, Gounopoulos, D, Platanakis, E, & Zhang, W. (2023). Operational
research and artificial intelligence methods in banking. European Journal of Operational Research,
306(1), 1-16.
21.
Easwaran, V., Orayj, K., Goruntla, N., Mekala, J. S., Bommireddy, B. R., Mopuri, B., ... & Bandaru, V.
(2025). Depression, anxiety, and stress among HIV-positive pregnant women during the COVID-19
pandemic: a hospital-based cross-sectional study in India. BMC Pregnancy and Childbirth, 25(1), 134.
22.
Ekundayo, F, Atoyebi, I, Soyele, A, & Ogunwobi, E. (2024). Predictive Analytics for Cyber Threat
Intelligence in Fintech Using Big Data and Machine Learning. Int J Res Publ Rev, 5(11), 1-15.
23.
Fritz-Morgenthal, S, Hein, B, & Papenbrock, J. (2022). Financial risk management and explainable,
trustworthy, responsible AI. Frontiers in artificial intelligence, 5, 779799.
24.
Herrmann, H, & Masawi, B. (2022). Three and a half decades of artificial intelligence in banking,
financial services and insurance: A systematic evolutionary review. Strategic Change, 31(6), 549-569.
25.
Hood, K., & Al-Oun, M. (2014). Changing performance traditions and Bedouin identity in the North
Badiya, Jordan. Nomadic Peoples, 18(2), 78-99.
26.
Hsu, M. F, Hsin, Y. S, & Shiue, F. J. (2022). Business analytics for corporate risk management and
performance improvement. Annals of Operations Research, 1-41.
27.
Islam, T, Islam, S. M, Sarkar, A, Obaidur, A, Khan, R, Paul, R, & Bari, M. S. (2024). Artificial Intelligence
in Fraud Detection and Financial Risk Mitigation: Future Directions and Business Applications.
International Journal for Multidisciplinary Research.
28.
Jagdish Jangid. (2023). Enhancing Security and Efficiency in Wireless Mobile Networks through
Blockchain. International Journal of Intelligent Systems and Applications in Engineering, 11(4), 958
–
969. Retrieved from https://ijisae.org/index.php/IJISAE/article/view/7309
29.
Jagdish Jangid. (2023). Enhancing Security and Efficiency in Wireless Mobile Networks through
IJBMS, 2025 Page No. 05-14
IJBMS
29
Blockchain. International Journal of Intelligent Systems and Applications in Engineering, 11(4), 958
–
969. Retrieved from https://ijisae.org/index.php/IJISAE/article/view/7309
30.
Jaiswal, R. (2023). Impact of AI in the General Insurance underwriting factors. Central European
Management Journal, 697-705.
31.
Kalogiannidis, S, Kalfas, D, Papaevangelou, O, Giannarakis, G, & Chatzitheodoridis, F. (2024). The role
of artificial intelligence technology in predictive risk assessment for business continuity: A case study
of Greece. Risks, 12(2), 19.
32.
Kaniz, R. E., Lindon, A. R., Rahman, M. A., Hasan, M. A., & Hossain, A. (2025). The Impact of Project
Management Strategies on the Effectiveness of Digital Marketing Analytics for Start-up Growth in
the United States. Inverge Journal of Social Sciences, 4(1), 8-24.
33.
Kannan, N. (2024). The Role of Artificial Intelligence and Machine Learning in Personalizing Financial
Services in Banking and Insurance. International Journal of Banking and Insurance Management
(IJBIM), 2(1), 1-13.
34.
Kuppan, K, Acharya, D. B, & Divya, B. (2024). Foundational AI in Insurance and Real Estate: A Survey
of Applications, Challenges and Future Directions. IEEE Access.
35.
Mohammed, A. B, Al-Okaily, M, Qasim, D, & Al-Majali, M. K. (2024). Towards an understanding of
business intelligence and analytics usage: evidence from the banking industry. International Journal
of Information Management Data Insights, 4(1), 100215.
36.
Muhammad Zaurez Afshar1*, Dr. Mutahir Hussain Shah2. (2025). Strategic Evaluation Using PESTLE
and SWOT Frameworks: Public Sector Perspective. ISRG Journal of Economics, Business &
Management (ISRGJEBM), III(I), 108
–
114. https://doi.org/10.5281/zenodo.14854362
37.
Mullins, M, Holland, C. P, & Cunneen, M. (2021). Creating ethics guidelines for artificial intelligence
and big data analytics customers: The case of the consumer European insurance market. Patterns,
2(10).
38.
Nguyen, H. U., Trinh, T. X., Duong, K. H., & Tran, V. H. (2018). Effectiveness of green muscardine
fungus Metarhizium anisopliae and some insecticides on lesser coconut weevil Diocalandra frumenti
Fabricius (Coleoptera: Curculionidae). CTU Journal of Innovation and Sustainable Development,
(10), 1-7.
39.
Nguyen, L., Trinh, X. T., Trinh, H., Tran, D. H., & Nguyen, C. (2018). BWTaligner: a genome short-read
aligner. Vietnam Journal of Science, Technology and Engineering, 60(2), 73-77.
40.
Nwaimo, C. S, Adewumi, A, & Ajiga, D. (2022). Advanced data analytics and business intelligence:
Building resilience in risk management. International Journal of Scientific Research and Applications,
6(2), 121.
41.
Oyedokun, O, Ewim, S. E, & Oyeyemi, O. P. (2024). Leveraging advanced financial analytics for
predictive risk management and strategic decision-making in global markets. Global Journal of
Research in Multidisciplinary Studies, 2(02), 016-026.
42.
Pattnaik, D, Ray, S, & Raman, R. (2024). Applications of artificial intelligence and machine learning in
the financial services industry: A bibliometric review. Heliyon, 10(1).
43.
Paul, L. R, Sadath, L, & Madana, A. (2021). Artificial intelligence in predictive analysis of insurance and
banking. In Artificial Intelligence (pp. 31-54). CRC Press.
44.
Rahman, S., Sayem, A., Alve, S. E., Islam, M. S., Islam, M. M., Ahmed, A., & Kamruzzaman, M. (2024).
The role of AI, big data and predictive analytics in mitigating unemployment insurance fraud.
International Journal of Business Ecosystem & Strategy (2687-2293), 6(4), 253-270.
45.
Rahmani, F. M, & Zohuri, B. (2023). The transformative impact of AI on financial institutions, with a
focus on banking. Journal of Engineering and Applied Sciences Technology. SRC/JEAST-279. DOI:
doi. org/10.47363/JEAST/2023 (5), 192, 2-6.
46.
Rasul, I., Akter, T., Akter, S., Eshra, S. A., & Hossain, A. (2025). AI-Driven Business Analytics for
IJBMS, 2025 Page No. 05-14
IJBMS
30
Product Development: A Survey of Techniques and Outcomes in the Tech Industry. Frontline
Marketing, Management and Economics Journal, 5(01), 16-38.
47.
Sachin Dixit, & Jagdish Jangid. (2024). Asynchronous SCIM Profile for Security Event Tokens. Journal
of Computational Analysis and Applications (JoCAAA), 33(06), 1357
–
1371. Retrieved from
https://eudoxuspress.com/index.php/pub/article/view/1935
48.
Sachin Dixit, & Jagdish Jangid. (2024). Asynchronous SCIM Profile for Security Event Tokens. Journal
of Computational Analysis and Applications (JoCAAA), 33(06), 1357
–
1371. Retrieved from
https://eudoxuspress.com/index.php/pub/article/view/1935
49.
Sawyer, S., Ellers, S., Kakumanu, M. S., Bommireddy, B., Pasgar, M., Susan-Kurian, D., ... & Jurdi, A.
A.
(2025).
Trial
in
progress
for
a
colorectal
cancer
detection
blood
test.
https://ascopubs.org/doi/10.1200/JCO.2025.43.4_suppl.TPS306
50.
Zaurez Afshar, M., & Hussain Shah, M. (2025). Performance Evaluation Using Balanced Scorecard
Framework: Insights from A Public Sector Case Study. INTERNATIONAL JOURNAL OF HUMAN AND
SOCIETY, 5(01), 40
–
47. https://ijhs.com.pk/index.php/IJHS/article/view/808
51.
Zhao, Y. (2024). Integrating advanced technologies in financial risk management: A comprehensive
analysis. Advances in Economics, Management and Political Sciences, 108, 92-97.
52.
Zhou, D, & Zhang, Z. (2024). AI Applications for Financial Risk Assessment: The Future of Predictive
Analytics in the Financial Sector. Journal of Financial Innovations, 18(2), 237-256.
