See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/382276460

"Impact of Artificial Intelligence on Financial reporting: Evidence from KPMG

and S & P GLOBAL"

Thesis · July 2024

DOI: 10.13140/RG.2.2.22064.78089

CITATIONS READS

0 1,006

3 authors:

Qazi TAHA FATEEN Lodhi Muhammad Ossama Khan

Bahria University Islamabad Campus Bahria University Islamabad Campus
1 PUBLICATION 0 CITATIONS 1 PUBLICATION 0 CITATIONS

SEE PROFILE SEE PROFILE

Shanzay Adnan
Bahria University
1 PUBLICATION 0 CITATIONS

SEE PROFILE

All content following this page was uploaded by Qazi TAHA FATEEN Lodhi on 16 July 2024.

The user has requested enhancement of the downloaded file.

“Impact of Artificial Intelligence on Financial reporting: Evidence from KPMG and
S & P GLOBAL”

By:

Muhammad Ossama Khan 01-112202-077

Qazi Taha Fateen Lodhi 01-112202-107
Shanzay Adnan 01-112202-092

A&F
Supervisor
Sir Ahmed Bilal Siddiqui

Department of Management Studies

Bahria Business School
Bahria University Islamabad
Spring 2024

1
Contents
Key words: ................................................................................................................................................... 3
Introduction: ............................................................................................................................................... 3
Impact of Artificial Intelligence on Financial Reporting: ....................................................................... 3
Introduction to AI in Financial Reporting........................................................................................ 3
Applications of AI in Financial Reporting ........................................................................................ 4
Benefits of AI in Financial Reporting ............................................................................................... 5
Challenges in Implementing AI in Financial Reporting .................................................................. 6
Case Studies and Real-World Examples ........................................................................................... 7
Future of AI in Financial Reporting ................................................................................................. 7
Ethical Considerations and Best Practices ....................................................................................... 8
Objectives of the study: .............................................................................................................................. 9
Research Hypotheses: ............................................................................................................................. 9
Literature review ...................................................................................................................................... 10
Conceptual Review: .............................................................................................................................. 10
Financial Reporting: ......................................................................................................................... 10
Financial Reporting process (FRP): ................................................................................................ 10
Artificial Intelligence: ....................................................................................................................... 10
Machine Learning (ML):.................................................................................................................. 11
Cloud Accounting: ............................................................................................................................ 11
Overview of KPMG in Pakistan: ..................................................................................................... 12
Overview of S and P Global in Pakistan: ........................................................................................ 12
Theoretical examination:...................................................................................................................... 12
Task-Technology fit theory: ................................................................................................................. 12
Diffusion of Innovation (DOI) Theory, ........................................................................................... 13
Empirical Review:................................................................................................................................. 14
Conceptual framework:............................................................................................................................ 15
Methodology: ......................................................................................................................................... 15
Validity and Reliability of the Instrument: ........................................................................................ 16
Reliability Statistics ................................................................................................................................ 16
Reliability Analysis for each variable: ................................................................................................ 16
The Correlation between AI and other variables: ......................................................................... 19
 Correlation between ML and other variables: ............................................................................... 19
Correlation between FRP and other variables:.............................................................................. 20
Correlation between FRQC and other variables: .......................................................................... 20
Significance Levels: ........................................................................................................................... 20
Descriptive Statistics ............................................................................................................................. 20
Descriptive Statistics (Frequency): .......................................................................................................... 22
Artificial Intelligence(AI) ..................................................................................................................... 22
Statistics .............................................................................................................................................. 22
Machine learning (ML): ....................................................................................................................... 24
Financial Reporting Process(FRP): ..................................................................................................... 25
Financial Reporting Qualitative data (FRQC): ................................................................................. 27
Regression .................................................................................................................................................. 29
Machine Learning (ML) → Financial Reporting Process (FRP) ..................................................... 29
Regression results ............................................................................................................................... 29
ANOVA .............................................................................................................................................. 30
Coefficients ......................................................................................................................................... 30
Regression Results .............................................................................................................................. 31
Test of Hypotheses .................................................................................................................................... 33
Model Summaryb .................................................................................................................................... 33
Conclusion: ................................................................................................................................................ 34
Model Summary.......................................................................................................................................... 35
Discussion of findings: .............................................................................................................................. 36
Recommendations:.................................................................................................................................... 38
Assimilation of Machine Learning Technology (ML): ...................................................................... 39
Investment in Information Technology Infrastructure: .................................................................... 39
Integration of Machine and Deep Learning Algorithms: .................................................................. 39
Conclusion: ................................................................................................................................................ 40
References:................................................................................................................................................. 42

1
 Impact of Artificial Intelligence on Financial reporting: Evidence from
KPMG and S & P GLOBAL

Financial Reporting: To demonstrate a company's financial status and performance, financial

reports include balance sheets, cash flow statements, and income statements.
Artificial intelligence (AI): AI is the ability of a computer system to carry out tasks that would
typically require intelligence from humans, such as speech recognition, pattern recognition,
decision-making, language translation, and language understanding.
Machine learning (ML): A computer programming language, where machines directly
corresponds to the instructions given by computer hardware to improve their performance over
time. It includes tasks like recognizing patterns and predicting outcomes. Machine learning is
basically a method to achieve AI by allowing machines to learn from data.

Overview:
The Financial reporting can be sometimes so complex, high priced, with possibilities of material
misstatement, to tackle this, study evaluated how artificial intelligence (AI) could help us with
FRP. We constructed a questionnaire for the survey to collect data and assessed it using regression
analysis on SPSS software (IBM SPSS statistics). The study examined how effective AI and ML
is for financial reporting process and how it can affect the quality of financial reports. The results
show that AI and ML significantly enhances financial reporting process, The P value is less than
5% which means the relationship is statistically significant, indicating strong evidence. Taking
into account these results, the study determined that companies need to adopt Artificial intelligence
technologies to stay spirited and competitive. In order to increase productivity and efficiency, it

2
advised businesses to begin putting up information resources for artificial intelligence (AI), such
as networks, AI software, instruments for information annotation, data refinement, storage of
information and security, and networks.

Key words:
Machine learning (ML), Artificial Intelligence (AI), Financial reporting process (FRP), Financial
reporting Qualitative Characteristics (FRQC), Machine-to-machine communication, Internet of
things.

Introduction:
The Fourth Industrial Revolution (4IR) is the current and developing environment in which trends
and technologies such as machine-to-machine communication (M2M), the internet of things (IoT),
virtual reality (VR), robotics and AI are changing people living standards and work life. It has
initiated a new era characterized by advancement in technology, among which AI stand out on the
top, In the field of accounting and finance, the fusion of AI technologies is increasingly prevailing
everywhere specially in firms and organizations, offering solutions to the root problems and
challenges of data processing, efficiency and accuracy. Evidently, foremost firms such as S&P
global and KPMG have enveloped AI to streamline their financial reporting processes and enhance
their decision-making capabilities.
AI technologies span a range of techniques and tools, including machine learning, deep learning,
natural language processing, cloud accounting and cloud computing and development. By
leveraging AI, accounting firms can improve the speed, accuracy, and reliability of financial
reporting, ultimately enhancing the quality of information available to stakeholders. These AI
technologies enable automation, data analysis, and pattern recognition, thereby transforming
traditional accounting practices.

Impact of Artificial Intelligence on Financial Reporting:

1. Introduction to AI in Financial Reporting
Financial reporting is just one of the industries that artificial intelligence (AI) is transforming. Its
application in this field improves compliance, efficiency, and accuracy. Financial reporting
processes can be automated and improved with the help of artificial intelligence (AI), which

3
includes a range of technologies such as machine learning (ML), natural language processing
(NLP), and robotic process automation (RPA). Financial reports are now much more efficient and
of higher quality thanks to the data processing, analysis, forecast, and decision-making capabilities
of these technologies.

2. Applications of AI in Financial Reporting

a. Automated Data Entry and Processing
AI systems can automate the collection and processing of data from a variety of sources, including
bank statements, invoices, and receipts. These systems, in particular, may do so by leveraging
RPA. These systems reduce manual mistake and speed up the process by reliably capturing data
through the use of established rules and learning algorithms. An AI-powered system, for instance,
can scan an invoice, extract the pertinent financial data, and enter it into the applicable accounting
software fields.
b. Natural Language Processing (NLP) for Report Generation
AI can produce and understand human language thanks to NLP. NLP can be used in financial
reporting to create narrative reports from structured data. For example, by evaluating financial data
and creating summaries, AI technologies can automatically prepare accounting records and
management reports, saving accountants a significant amount of time. One instance is when AI is
used to create annual reports, summarizing financial performance and emphasizing important
metrics and patterns.
c. Anomaly Detection and Fraud Prevention
Large datasets can be analyzed by machine learning algorithms, which can then look for trends
and abnormalities that might point to fraud. By continuously learning from past data, these
algorithms are able to spot inconsistencies that human auditors might overlook, improving the
accuracy of financial reporting. AI, for example, has the ability to track transactions in real-time
and identify questionable behavior, such as unusually high costs or irregular vendor payments.
d. Predictive Analytics
Predictive analytics is used by AI to project financial results based on past data. This capacity is
essential for risk management, financial planning, and budgeting. Organizations can make wise
decisions by using predictive models, which can offer insights into future income, expenses, and

4
other important financial variables. An AI system may, for instance, examine historical sales data
to forecast future sales trends, enabling companies to make proactive strategy adjustments.

e. Compliance and Regulatory Reporting

By automatically maintaining and updating records in accordance with the most recent regulations,
AI helps to ensure compliance with regulatory standards. It can also provide reports that regulatory
bodies need, guaranteeing timeliness and accuracy. AI, for example, may automate the creation of
tax reports and guarantee that they adhere to the most recent tax laws and guidelines.

3. Benefits of AI in Financial Reporting

a. Increased Efficiency and Productivity
AI frees up crucial time for financial analysts to concentrate on more strategic operations by
automating repetitive and routine chores. Increased productivity and quicker report output are the
results of this automation. For instance, with AI support, tasks that used to take days to collect and
confirm information can now be completed in a few hours.
b. Enhanced Accuracy and Reduced Errors
When entering and analyzing data, human error is greatly decreased by AI systems. AI guarantees
excellent precision in financial reports by continuously implementing established principles and
learning from data. In reporting on finances, where even minor mistakes can have serious
repercussions, accuracy is essential.
c. Cost Savings
AI lowers the necessity for manual labor by automating several financial reporting tasks, which
results in considerable cost savings. Redistributing resources to higher-value endeavors is an
option for organizations. The long-term cost reductions from lower personnel expenses and higher
productivity can more than offset the initial investments in AI technology.
d. Improved Compliance and Risk Management
Artificial Intelligence (AI) assists firms in adhering to changing regulatory standards by
continuously monitoring modifications and adjusting processes as necessary. Additionally, it
improves risk management by seeing possible problems before they become serious ones. For

5
example, AI can highlight potential compliance problems in financial transactions, enabling for
quick corrective steps.
e. Better Decision-Making
AI uses predictive modeling and sophisticated data analytics to deliver deeper insights. Finance
teams may make better decisions thanks to this information, which boosts overall business
performance. For example, by examining spending trends and recommending more effective
options, AI can assist in locating cost-saving potential.

4. Challenges in Implementing AI in Financial Reporting

a. Data Quality and Availability
AI systems depend significantly on well-organized, high-quality data. It can be difficult to
guarantee that clean, correct data is available, especially for businesses that use older systems and
several data sources. Inconsistent data formats and data silos can reduce the efficacy of AI
applications.
b. Integration with Existing Systems
AI integration with current financial operations and systems can be difficult and time-consuming.
Businesses must guarantee that AI solutions work in unison with their present IT setup. This could
need making big, expensive, and disruptive changes to the current systems and operations.
c. Skills and Expertise
Financing, machine learning, and data science expertise are necessary for the successful
application of AI. Professionals that can cross these disciplinary boundaries are becoming more
and more in demand. To cultivate these talents internally or employ outside experts, organizations
may need to make training and development investments.
d. Cost and Resource Allocation
Even though AI might result in long-term cost benefits, there may be a substantial upfront cost for
AI infrastructure and technology. Companies must give serious thought to how they will allocate
their resources and money. This covers the price of the technology itself as well as the resources
required for setup and continuing upkeep.
e. Ethical and Regulatory Concerns

6
Concerns about confidentiality of information and algorithmic transparency are among the ethical
and legal issues that are brought up by the adoption of AI in reporting on finances. To preserve
confidence and guarantee compliance, organizations need to overcome these obstacles. For
instance, sustaining stakeholder confidence depends on making sure AI systems do not
unintentionally introduce bias or make conclusions that are difficult to explain.

5. Case Studies and Real-World Examples

a. JPMorgan Chase
JPMorgan Chase has employed a means called COIN (Contract Intelligence), which leverages
artificial intelligence, to analyze legal papers and extract significant information. Legal teams used
to need hundreds of hours to analyze thousands of papers; this system can do the same process in
seconds. COIN has improved accuracy while drastically cutting down on the time and expense of
document inspection.
b. Deloitte
Deloitte analyzes complicated contracts for clients using an AI program called Argus. Deloitte is
able to offer more thorough and timely consulting services thanks to Argus's ability to swiftly
detect important terms, potential threats, and compliance issues in contracts. This application
effectively interprets and understands legal text by utilizing ML and NLP.
c. Ernst & Young (EY)
Ernst and Young has created an AI audit platform called EY Canvas, which automates repetitive
processes and conducts extensive data analysis to improve the audit process. AI is used by EY
Canvas to find anomalies and possible threats in financial accounts, enhancing the accuracy and
dependability of audits. Additionally, this solution has improved the efficiency of the audit
procedure by streamlining it.

6. Future of AI in Financial Reporting

a. Advanced Predictive and Prescriptive Analytics
We may anticipate increasingly advanced prescriptive and predictive analytics capabilities as AI
technology develops. Even more precise projections and useful perspectives for financial decision-
making and planning will be made possible by these developments. Prescriptive analytics will

7
provide customized recommendations based on several circumstances, while predictive models
will continue to improve in accuracy.
b. Greater Automation and Integration
More automation capabilities from AI systems in the future will further minimize the requirement
for human interaction in mundane jobs. A more comprehensive approach to finance management
will be made possible by improved integration with other company systems. AI might, for
example, easily interface with ERP (enterprise resource planning) systems to deliver financial
insights in real time.
c. Real-Time Reporting and Analysis
Directly analysis and reporting of finances made possible by AI will give businesses the most
recent information about their financial performance. This skill will be essential for making quick
decisions in a hectic work setting. Executives will be able to keep an eye on important data and
act fast in response to new trends thanks to real-time dashboards driven by AI.
d. AI-Driven Audits
AI will be used increasingly frequently in auditing, helping to improve the precision and
effectiveness of audits by automating various parts of the process. artificial intelligence-driven
audits will be less likely to contain human error and more comprehensive. Auditors can concentrate
on regions with high risk while AI takes care of regular data verification and analysis.
e. Personalized Financial Insights
AI will offer more individualized financial insights based on the unique requirements and
objectives of different users and businesses. Financial reports will be more valuable and relevant
as a result of this customization. AI might, for instance, provide small company owners with
tailored financial advice based on their particular financial circumstances and market trends.

7. Ethical Considerations and Best Practices

a. Ensuring Transparency
Businesses need to make sure AI algorithms are comprehensible and transparent. This entails
recording the decision-making process used by AI systems and making sure stakeholders can
understand the reasoning behind them. Maintaining confidence and adhering to regulations both
depend on transparency.

8
 b. Mitigating Bias
AI systems need to be created with fairness and bias reduction in mind. This entails utilizing a
variety of training data and constantly checking AI outputs for biased indicators. To handle such
biases and guarantee the ethical use of AI, organizations should have strong governance
frameworks in place.
c. Data Privacy and Security
Maintaining the safety of financial information and safeguarding data privacy are critical. To
protect sensitive data, organizations must adhere to data protection laws and put robust security
measures in place. This covers the use of access controls, encryption, and frequent security
assessments.
d. Ongoing Monitoring and Evaluation
To make sure artificial intelligence (AI) systems are operating as planned, they should be regularly
observed and assessed. This entails routine evaluations of AI results, performance indicators, and
adherence to moral guidelines. Companies ought to be ready to modify AI systems as necessary.

Objectives of the study:

The study seeks to address the following objectives:
1. To evaluate the effectiveness of machine learning in gathering data and identification of
anomalies in financial transactions, data compilation in financial dealings, information
acquisition of discrepancies in accounting entries.
2. To uncover the impact of AI in producing relevant income statements, financial statements
and cash flow statements that is understandable, similar and substantial promptly.

Research Hypotheses:
Based on the objectives of the study, we have devised the following null hypothesis:
Ho1: There is no significant effect of ML on data compilation, recording or identifying
anomalies/irregularities accounting or financial transactions.
Ho2: There is no significant effect of AI on financial reporting process (FRP) and financial
reporting Qualitative characteristics.

9
Literature review
Conceptual Review:
Financial Reporting:
In financial reporting we compose documents like balance sheet, cash flow
statements and income statements to show a company’s financial position and
performance. Financial reporting is about transmitting companies financial detail
and information with different type of controllers, like investors, managers and
directors. The mission is to help these users or controllers make decisions. Income
statement and financial position show details about company’s finances, loan, debts,
profits and losses. The financial statements should be easy to understand and give a
clear picture of company’s financial health. These statements need to have specific
attributes to be helpful.

Financial Reporting process (FRP):

The financial reporting process include various steps like identifying financial
transactions, recording financial transaction in journals and posting them to ledgers.
Trial balances are adjusted and prepared to guarantee accuracy before compiling
financial statements. With emerging technologies, the financial reporting process is
developing. New technologies allow more accurate and faster reporting to stay
ambitious organizations need to continuously improve and evaluate their processes.
Financial reporting process is imperative for making informed business decisions.
Having high-quality and timely financial information enables businesses to add
value and make better daily decisions.

Artificial Intelligence:
John McCarthy is considered as the father of Artificial Intelligence. John McCarthy
was an American computer scientist. The term "artificial intelligence" was coined
by him. Computer system that is able to perform tasks that normally requires human
intelligence, such as recognizing speech and patterns, and decisions-making,
translation between different languages and understanding languages. Furthermore,
artificial intelligence (AI) in 1956 with a convention at Dartmouth College.
Researcher there wanted to generate machines that could think like human beings

10
 but with lower risk of error and faster. In 1960s, they created expert systems that
could do tasks requiring human expertise. In following years 1970s and 1980s, they
focused on making machines and technologies that could learn and improve
themselves like humans do. Nowadays, AI is more efficient such that machines can
think learn and work from experiences. They can make better and accurate decisions
like experts in various fields. AI is used in different and many areas of life, and it
keeps getting better with advancements and technologies like deep learning,
machine to machine communications, machine learning and others.

Machine Learning (ML):

Machine learning history starts in 1943 with the first mathematical model of neural
networks. Machine learning is a technology that allows AI systems to learn, forget,
and adapt on their own. In accounting, machine learning is used for tasks like audit
investigations, bank reconciliation, and assessing credit worthiness. It includes
creating computer programs that can analyze data, recognize patterns and different
languages and.learn from it, to make decisions automatically. For example, in bank
reconciliation, machine learning algorithms study past transactions to understand
how they were categorized. They use this information to determine how to handle
current transactions. If a mistake was made in the past and corrected, the algorithm
remembers the correct entries to guide future actions.

Cloud Accounting:
Cloud Accounting operates similarly to self-install accounting software or
traditional on premises however accounting software is hosted on remote server
adopting a software as service business model. Data is set into the cloud process and
then return to the user. In contrast, cloud accounting is a modern approach that offers
accounting and reporting services over the internet. Traditional accounting software
is usually bought and installed on company servers or individual computers. AI
programs can capture and process information as it happens, and users can access
financial reports anytime for decision-making. With cloud accounting, you can
access your accounting database online anytime anywhere. Cloud accounting
software is hosted on a cloud server, making accounting information available 24/7.
This means regardless of location, transactions can be updated and recorded in real-

11
 time. This innovation saves companies the cost of maintaining an internal IT and
setting up IT system, allowing them to invest more in production while still getting
top-quality IT services through the cloud accounting.

Overview of KPMG in Pakistan:

KPMG in Pakistan embodies the value proposition of KPMG's global network in
providing professional services to organizations in both the public and private
sectors which rely on KPMG for the integrity of their audit opinion, the provision of
sound and valuable business guidance and advice on potential business risks. KPMG
is an international firm that offer audit, consultancy, tax and advisory services to
local and international organizations. Each firm under the umbrella of KPMG
International Limited and are separate legal entity within their jurisdiction.
Furthermore, KPMG operate in 155 countries and have more than 174,000 people
working in member firms around the world. The KPMG network is coordinated by
Swiss cooperative KPMG International (KPMGI). KPMG Taseer Hadi & Co., was
established in 1969 in Pakistan with two Partners, awarded “Representative Firm”
status of Peat Marwick in 1973, “Correspondence Firm Status” in 1977 and a full
“Member Firm” status of Peat Marwick in 1982.

Overview of S and P Global in Pakistan:

S&P Global roots stretch back to 1860 with a vision for how Essential Intelligence
Could Accelerate Progress in the industrial revolution. Integrity, discovery and
partnership are embedded into our identity. S and P Global works constantly to
expand our data and insight capabilities. In an ever-changing world, our transparent,
trusted solutions provide confidence for critical decisions. Customers worldwide
depend on S and P Global data and insights. We’ve become the global standard in
providing critical financial information to our customers. S and P Global is a firm
that offer audit, consultancy, tax and advisory services.

Theoretical examination:
Task-Technology fit theory:
Task-Technology Fit (TTF), as defined by Goodhue and Thompson (1995), is the
appropriateness of implementing a new technology to carry out a task. An example

12
 of this would be the use of artificial intelligence (AI) and its sibling technologies in
the input, processing, and output stages of the financial reporting and accounting
process. According to the concept, a technology is considered task-fit if its features
match those of the task and its users anticipate that using it will improve
performance. While the AI technology has characteristics like speed, accuracy,
meticulousness, versatility, reliability, memory, and consistency, the task
characteristics are those of financial reporting, including relevance, faithful
representation, and comparability, verifiability, timeliness, and understandability.
Users will be more inclined to employ artificial intelligence (AI) in financial
reporting if both of these traits match and they will be certain of the performance
benefit.
2.2.1 A structured questionnaire was employed as the study's principal data collection
method. OLS regression analysis was then used to examine the questionnaire
responses in order to ascertain the influence of artificial intelligence and machine
learning on financial reporting procedures as well as the qualitative aspects of these
processes. This indicates that in order to get precise information from respondents
regarding how artificial intelligence is affecting financial reporting, the researchers
created a set of predetermined questions. OLS regression analysis was then used to
examine the questionnaire responses in order to ascertain the influence of artificial
intelligence and machine learning on financial reporting procedures as well as the
qualitative aspects of these processes.

Diffusion of Innovation (DOI) Theory,

developed by E.M. Rogers in 1962, is one of the oldest social science theories. It
originated in communication to explain how, over time, an idea or product gains
momentum and diffuses (or spreads) through a specific population or social system.
This theory helps explain how new technologies, like AI, are accepted in fields such
as accounting and financial reporting. The Dol theory highlights five key
characteristics of innovations that influence their adoption: relative advantage,
observability, compatibility complexity and trainability. (Wani & Ali, 2015).

13
Empirical Review:
Our basis study states that Odoh et al. (2018) investigated how artificial intelligence
affected the efficiency of accounting procedures in southeast Nigerian accounting
businesses. 185 managers and accountants participated in the study, which used a
descriptive research approach. With the distribution of a standardized questionnaire,
primary data was employed. At the 5% level of significance, linear regression was used to
evaluate the study hypotheses. Expert system (r =.904; F = 608.447; p =.000) and
intelligent agent (r =.754a; F = 178.810; p =.000) were shown to have a considerable impact
on the accounting functions performed by accounting companies in South East Nigeria,
according to the study's findings. The impact of automation and cutting-edge technology
like machine learning on Liberia's financial reporting procedures was examined by Sylvia
and Etchi (2019). Of the four large firms (Deloitte, PwC, EY, and KPMG), two (names
withheld) used an exploratory study design. Using a purposeful nonprobability sampling
technique, the sample size by which the primary data were gathered was established. The
data was analyzed using thematic analysis technique. The study discovered that by
reducing errors, raising productivity and speed, and improving efficiency, automation
using AI technologies has enhanced the standard of financial reports. It was determined
that the reporting process has been improved by automation. The use of machine learning
(ML) and artificial intelligence (AI) in the Indian tax system was investigated by Rathi,
Sharma, and Lodha (2021). A structured questionnaire using a 5-point Likert scale—from
1 for strongly agreeing to 5 for strongly disagreeing—was used to collect primary data.
The Pearson product-moment correlation analysis was conducted to evaluate if the
application of AI and ML will simplify the tax system and improve its equity. According
to the study's findings, there is a strong correlation (p =.000) and a high favorable correction
(r =.817, n = 50) between the variables. In their 2021 study, Al-Sayyed, Al-Aroud, and
Zayed looked at how AI technologies affected audit evidence.
The study used a quantitative research design. Via the distribution of a standardized
questionnaire, primary data was employed. Regression analysis's inferential statistic was
employed to assess the study's hypotheses at a significance level of 5%, with the hypothesis
being rejected if the p-value is below 0.05. According to the study's findings, the expert

14
 system significantly affects audit evidence (r=.906, R square =.847, an F statistic =
709.457, t = 24.767; p < 0.05).

Conceptual framework:
The study employed Artificial Intelligence and Financial Reporting as variables, which were
represented by Financial Reporting Process (FRP), Financial Reporting Qualitative
Characteristics (FRQC), and Machine Learning (ML). The objective of the model is to assess
how well machine learning performs when it comes to data collection, anomaly detection in
financial transactions, financial data compilation, and information collecting on
inconsistencies in accounting entries. Additionally, to ascertain how AI affects the timely,
comparable, and significant production of pertinent income, financial, and cash flow
statements. The following conceptual framework, as developed by the researchers, served as
the basis for this investigation.

Independent Variables Dependent Variables

→
Machine learning (ML) Financial Reporting process(FRP)

Artificial Intelligence (AI) Financial Reporting Qualitative

Characteristics (FRQC)

Methodology:
We designed a questionnaire measured by 5-point a Likert, scale 1: Strongly Disagree 2: disagree,
3: Neutral, 4: Agree, 5: Strongly agree and then we conducted a survey with a total of 60
Statements (consisting of 15 Questions or statements for each independent and dependent variable)
from KPMG and S and P Global. The sample size or total responses we got on all the variables is
around 67-68. The data was collected through this structured questionnaire, and the analysis was
performed using the Ordinary Least Square (OLS) regression model with IBM SPSS version
29.0.2.0 (20).

A Pearson product-moment correlation was computed to assess if AI and ML usage will reduce
the complexity of tax system.

15
Correlation is significant at the 0.01 level (2-tailed).**
Correlation is significant at the 0.05 level (2-tailed).*

Validity and Reliability of the Instrument:

The questionnaire was subject to content validity, where it was properly scrutinized by experts in
the field of Accounting and Finance. who provided suggested modifications and feedback, which
were incorporated into the final version. To ensure the reliability of the research, the reliability
was measured using Cronbach's alpha, there are total 60 items of the variables and Cronbach alpha
value is 0.922, indicating that the instrument is 92.2% reliable and 7.8% unreliable. According to
Tavakol and Dennick (2011), a Cronbach's alpha between 0.70 and 0.95 is considered an
acceptable range for the reliability of research instruments.

Reliability Statistics
Cronbach's
Alpha N of Items
.922 60

There are total 60 items of the variables and Cronbach alpha value is 0.922 which is quite good because
Cronbach alpha value more than 0.7 is considered good, so in this case there is no issue of reliability.

Reliability Analysis for each variable:

1) AI

Reliability Statistics

16
Cronbach's
Alpha N of Items
.809 15

There are total 15 items of the variables and Cronbach alpha value is 0.809 which is quite good
because Cronbach alpha value more than 0.7 is considered good, so in this case there is no issue
of reliability.

2) ML

Reliability Statistics
Cronbach's
Alpha N of Items
.829 15

There are total 15 items of the variables and Cronbach alpha value is 0.829which is quite good
because Cronbach alpha value more than 0.7 is considered good, so in this case there is no issue
of reliability.

3) FRP

Reliability Statistics

Cronbach's
Alpha N of Items
.836 15

17
There are total 15 items of the variables and Cronbach alpha value is 0.836 which is quite good
because Cronbach alpha value more than 0.7 is considered good, so in this case there is no issue
of reliability.

4) FRQC

Reliability Statistics
Cronbach's
Alpha N of Items
.846 15

There are total 15 items of the variables and Cronbach alpha value is 0.846 which is quite good
because Cronbach alpha value more than 0.7 is considered good, so in this case there is no issue
of reliability.

Correlation

AI ML FRP FRQC
AI 1 .

ML .782** 1

FRP .397** .281* 1

FRQC .325** .317** .783** 1

18
Correlation is significant at the 0.01 level (2-tailed).**
Correlation is significant at the 0.05 level (2-tailed).*

The Correlation between AI and other variables:

• AI and ML: The correlation coefficient value between AI and ML is 0.782 which shows a
high positive association between both variables. The P value is < 0.001 which means the
relationship is statistically significant. So we can say that increase in level of AI can also
increase the ML. (r = 0.782, p < 0.001, N = 68)

• AI and FRP: The correlation coefficient value between AI and FRP is 0.397 which shows a
low positive correlation between both variables. The P value is < 0.001 which means the
relationship is statistically significant. So we can say that increase in level of AI can also
increase FRP. (r = 0.397, p < 0.001, N = 68)

• AI and FRQC: The correlation coefficient value between AI and FRQC is 0.325 which shows
a low positive association between both variables. The P value is 0.007 which is less than 0.05,
which means the relationship is statistically significant. So we can say that increase in level of
AI can also increase FRQC. (r = 0.325, p = 0.007, N = 67)

Correlation between ML and other variables:

• ML and AI: The correlation coefficient value between ML and AI is 0.782 which shows a
Strong positive association between both variables. The P value is < 0.001 which is less than
0.05 means the relationship is statistically significant. So we can say that increase in level of
ML can also increase AI. (r = 0.782, p < 0.001, N = 68)

• ML and FRP: The correlation coefficient value between ML and FRP is 0.281 which shows
a Weak positive correlation between both variables. The P value is 0.020 which is less than
0.05 means the relationship is statistically significant. So we can say that increase in level of
ML can also increase FRP (r = 0.281, p = 0.020, N = 68)

19
• ML and FRQC: The correlation coefficient value between ML and FRQC is 0.317 which
shows a Low positive correlation. The P value is 0.009 which is less than 0.05 means the
relationship is statistically significant. (r = 0.317, p = 0.009, N = 67)

Correlation between FRP and other variables:

• FRP and AI: Low positive correlation. The relationship is statistically significant and the
number of observations used in the calculation of the correlation coefficient are N= 68. (r =
0.397, p < 0.001, N = 68)

• FRP and ML: Weak positive correlation and is statistically significant. (r = 0.281, p = 0.020,
N = 68)

• FRP and FRQC: Strong positive correlation and is statistically significant (r = 0.783, p <
0.001, N = 67).

Correlation between FRQC and other variables:

• FRQC and AI: Low positive correlation. P value is 0.007 means the relationship is statistically
significant. (r = 0.325, p = 0.007, N = 67)

• FRQC and ML: Low positive correlation and is statistically significant. (r = 0.317, p = 0.009,
N = 67)

• FRQC and FRP: High positive correlation (r = 0.783, p < 0.001, N = 67)

Significance Levels:
• p < 0.01: Correlation is significant at the 0.01 level (2-tailed).

• p < 0.05: Correlation is significant at the 0.05 level (2-tailed).

Descriptive Statistics
N Minimum Maximum Mean Std. Deviation
AI 68 38.00 75.00 59.5588 5.89556
ML 68 39.00 75.00 61.5441 5.84484
FRP 68 45.00 75.00 62.5588 5.74685

20
FRQC 67 45.00 75.00 63.1343 5.62703
Valid N (listwise) 67

1. Sample Size (N): The total responses we got on all the variables is around 67-68.
2. Range of Values:
• AI: The minimum response we got on AI questions is 38.00 (means we got this much
strongly disagreed or disagreed responses from the people on the AI Qs) and the
maximum response we got on AI Qs is 75.00(strongly agreed or agreed to the Qs).
• ML: The minimum response we got on ML questions is 39.00 and the maximum
response we got on ML Qs is 75.00.
• FRP: The minimum response we got on FRP questions is 45.00 and the maximum
response we got on FRP Qs is 75.00.
• FRQC: The minimum response we got on FRQC questions is 45.00 and the
maximum response we got on FRQC Qs is 75.00.
3. Mean Value:
• AI: The average value is 59.5588, so to extent we can say that the AI practices are
prevailing in the organizations and the respondents are of the view that AI concept is
spreading in the organizations,
• ML: The average value is 61.5441, we can say that the ML concepts are prevailing
in the organizations and people are known to this concept.
• FRP: The average value is 62.5588, the respondents are of the view that FRP concept
is spreading in the organizations.
• FRQC: The average value is 63.1343, so to very extent we can say that the FRQC
practices are emerging in the organizations and people are well known to this
practice.
4. Standard Deviation (Spread out)
• AI: The variation among AI response is 5.89556
• ML: The variation among ML response is 5.84484
• FRP: The variation among FRP response is 5.74685
• FRQC: The variation among FRQC response is 5.62703

21
Descriptive Statistics (Frequency):
Artificial Intelligence(AI)

Statistics
AI
N Valid 68
Missing 0
Mean 59.5588
Mode 57.00
Minimum 38.00
Maximum 75.00

AI
N %
38.00 1 1.5%
45.00 1 1.5%
48.00 1 1.5%
49.00 1 1.5%
51.00 1 1.5%
53.00 3 4.4%
54.00 2 2.9%
55.00 1 1.5%
56.00 4 5.9%
57.00 8 11.8%
58.00 5 7.4%
59.00 6 8.8%
60.00 7 10.3%

22
61.00 3 4.4%
62.00 2 2.9%
63.00 3 4.4%
64.00 5 7.4%
65.00 5 7.4%
66.00 2 2.9%
67.00 4 5.9%
68.00 1 1.5%
69.00 1 1.5%
75.00 1 1.5%

23
Machine learning (ML):

Statistics
ML
N Valid 68
Missing 0
Mean 61.5441
Mode 60.00
Minimum 39.00
Maximum 75.00

ML
N %
39.00 1 1.5%
45.00 1 1.5%
46.00 1 1.5%
51.00 1 1.5%
52.00 1 1.5%
53.00 1 1.5%
57.00 2 2.9%
58.00 1 1.5%
59.00 4 5.9%
60.00 14 20.6%
61.00 8 11.8%
62.00 5 7.4%
63.00 7 10.3%
64.00 5 7.4%
65.00 1 1.5%

24
66.00 2 2.9%
67.00 7 10.3%
68.00 1 1.5%
69.00 1 1.5%
70.00 1 1.5%
72.00 2 2.9%
75.00 1 1.5%

Financial Reporting Process(FRP):

Statistics
FRP
N Valid 68
Missing 0

25
Mean 62.5588
Mode 60.00
Minimum 45.00
Maximum 75.00

FRP
N %
45.00 1 1.5%
48.00 1 1.5%
50.00 1 1.5%
55.00 2 2.9%
58.00 3 4.4%
59.00 7 10.3%
60.00 18 26.5%
61.00 1 1.5%
62.00 7 10.3%
63.00 3 4.4%
64.00 1 1.5%
65.00 3 4.4%
66.00 3 4.4%
67.00 2 2.9%
68.00 4 5.9%
69.00 3 4.4%
70.00 1 1.5%
71.00 1 1.5%
72.00 2 2.9%
73.00 2 2.9%
74.00 1 1.5%

26
75.00 1 1.5%

Financial Reporting Qualitative data (FRQC):

Statistics
FRQC
N Valid 68
Missing 0
Mean 63.0882
Mode 60.00
Minimum 45.00
Maximum 75.00

FRQC

27
 N %
45.00 1 1.5%
51.00 1 1.5%
56.00 2 2.9%
57.00 2 2.9%
58.00 2 2.9%
59.00 7 10.3%
60.00 11 16.2%
61.00 7 10.3%
62.00 4 5.9%
63.00 2 2.9%
64.00 6 8.8%
65.00 3 4.4%
66.00 5 7.4%
68.00 2 2.9%
69.00 2 2.9%
70.00 2 2.9%
71.00 3 4.4%
72.00 2 2.9%
73.00 1 1.5%
75.00 3 4.4%

28
Regression

Machine Learning (ML) → Financial Reporting Process (FRP)

Table no:1
Regression results
Std. Error of the
Model R R Square Adjusted R Square Estimate
1 .281a .079 .065 5.55698
a. Predictors: (Constant), ML
As indicated in table no 1, we can see that R-square value is 0.079, which means that our exogenous
variable i.e. ML causes 7.9% change in the endogenous variable i.e. FRP.

29
Table no:2
ANOVA
Model Sum of Squares df Mean Square F Sig.
1 Regression 174.686 1 174.686 5.657 .020
Residual 2038.079 66 30.880
Total 2212.765 67
a. Dependent Variable: FRP
b. Predictors: (Constant), ML
The table no 2, ANOVA results shows that the P value is 0.020 which is less than 0.05. Hence
we can say that there is a significant relationship between our independent variable i.e.ML
and dependent variable i.e. FRP.

Table no:3
Coefficients
Standardized
Unstandardized Coefficients Coefficients
Model B Std. Error Beta t Sig.
1 (Constant) 45.557 7.180 6.345 <.001
ML .276 .116 .281 2.378 .020
a. Dependent Variable: FRP
The table no 3 shows the coefficients results. As indicated that the beta value is 0.281, which
means that the change in independent variable i.e.ML by one unit will bring about the change in
the dependent variable i.e. FRP by 0.281 units.
Furthermore, the beta value is positive, which indicates the positive relationship between ML and
FRP. Or in other words we can say that when ML increases by one unit the FRP will also increase
by 0.281 units.

30
 1) Artificial Intelligence (AI) → Financial Reporting Qualitative Characteristics
(FRQC)

Tables no:1
Regression Results
Std. Error of the
Model R R Square Adjusted R Square Estimate
1 .310a .096 .082 5.36204
a. Predictors: (Constant), AI
As indicated in table no 1, we can see that R-square value is 0.096, which means that our exogenous
variable i.e. AI causes 9.6% change in the endogenous variable i.e. FRQC.

31
Table no:2
ANOVA
Model Sum of Squares df Mean Square F Sig.
1 Regression 201.872 1 201.872 7.021 .010
Residual 1897.599 66 28.751
Total 2099.471 67
a. Dependent Variable: FRQC
b. Predictors: (Constant), AI
The table no 2, ANOVA results shows that the P value is 0.010 which is less than 0.05. Hence we
can say that there is a significant relationship between our independent variable i.e.AI and
dependent variable i.e. FRQC.

Table no:3
Coefficients
Standardized
Unstandardized Coefficients Coefficients
Model B Std. Error Beta t Sig.
1 (Constant) 45.553 6.650 6.850 <.001
AI .294 .111 .310 2.650 .010
a. Dependent Variable: FRQC
The table no 3 shows the coefficients results. As indicated that the beta value is 0.310, which
means that the change in independent variable i.e.AI by one unit will bring about the change in the
dependent variable i.e. FRQC by 0.310 units.
Furthermore, the beta value is positive, which indicates the positive relationship between AI and
FRQC. Or in other words we can say that when ML increases by one unit the FRP will also increase
by 0.310 units.

32
Test of Hypotheses
1. Hypothesis: Machine Learning is not Effectively Significant in data compilation,
recording or identifying anomalies/irregularities accounting or financial
transactions.

Model Summaryb
Adjusted R Std. Error of the
Model R R Square Square Estimate Durbin-Watson
1 .281a .079 .065 5.55698 1.677

33
a. Predictors: (Constant), ML
b. Dependent Variable: FRP

Based on the above data analysis, several key points indicates that this hypothesis should be
rejected, following are key statistics:
• R Value: 0.281
• R Square: 0.079
• Adjusted R Square: 0.065
• Standard Error of the Estimate: 5.55698
• Durbin-Watson Statistic: Durbin Watson Statistic: The value is 1.677, which suggests
that there is no significant autocorrelation in the residuals of the regression model. This
means that serial autocorrelation does not present a problem in the analysis.
The R Square value of 7.9% indicates that approximately 7.9% of the variance in the dependent
variable (FRP) is explained by the predictor or independent variable (ML).
The Adjusted R Square value, which takes into account the number of predictors in the model, is
slightly lower at 6.5%.
The Standard Error of the Estimate provides an estimate of the standard deviation of the errors in
the prediction of FRP.
The Durbin-Watson statistic of 1.677 suggests that there is no significant autocorrelation in the
residuals of the regression model.

Conclusion:
Given that the p-value is less than 0.05, the null hypothesis that Machine learning (ML)
is not effectively significant in data compilation, recording or identifying
anomalies/irregularities accounting or financial transactions not effectively significant,
should be rejected. The analysis supports that machine learning is indeed effectively
significant in these areas. This breakdown simplifies the statistical details and presents
a clear conclusion based on the provided table.

34
Model Summary
Adjusted R Std. Error of the
Model R R Square Square Estimate Durbin-Watson
1 .310a .096 .082 5.36204 1.709
a. Predictors: (Constant), AI
b. Dependent Variable: FRQC

2. Hypothesis: Artificial intelligence is not effectively significant in producing a

relevant income statements, financial statements and cash flow statements that is
understandable, similar and substantial promptly.

Based on the above data analysis, key points indicate that this hypothesis should be rejected:

1. Durbin Watson Statistic: The value is 1.709, which suggests that there is no significant
autocorrelation in the residuals of the regression model. This means that serial
autocorrelation does not present a problem in the analysis.
2. Model Summary:
• R Value: 0.310
• R Square: 0.096
• Adjusted R Square: 0.082
• Standard Error of the Estimate: 5.36204
• Durbin-Watson: 1.709
The R Square of 9.6% indicates that the model explains approximately 9.6% of the
variation in the dependent variable (Financial Reporting Quality Characteristics FRQC)
due to the predictor or independent variable (AI).
3. Regression Coefficients:
• The regression coefficient (β1) of Financial Reporting Process (FRP) with respect
to Machine Learning (ML) is 0.593. This implies that for every 1% increase in ML,
the FRP increases by approximately 0.593%.

35
 4. Statistical Significance:
• The p-value for both FRP and ML is 0.000, which is less than 0.05. This indicates
that the variables are statistically significant.
Conclusion:
Given that the p-value is less than 0.05, the null hypothesis that Artificial intelligence is
not effectively significant in producing a relevant income statements, financial statements
and cash flow statements that is understandable, similar and substantial promptly, should
be rejected. The analysis supports that Artificial Intelligence is indeed effectively
significant in these areas. This breakdown simplifies the statistical details and presents a
clear conclusion based on the provided table.

Discussion of findings:
The statistical analysis of the two hypotheses uncovered that the significance of each tested
hypothesis is below the 0.05 decision criterion. This indicates that machine learning (ML) has
a statistically significant effect on the financial reporting process. ML's ability to access data,
learn from it, work on it and detect different patterns autonomously enables it to annotate
correct entries based on past corrections, thus enhancing the accuracy of financial postings.
These findings align with Sylvia and Etchi's (2019) study, which emphasized the effectiveness
of emerging technologies like machine learning in transaction classification, analysis, and
recording.
Furthermore, the study found that artificial intelligence (AI) significantly impacts the
qualitative characteristics of financial reporting. This suggests that AI facilitates the production
of relevant, faithfully represented financial statements that are comparable, verifiable, and
understandable in a timely manner, surpassing the capabilities of human intelligence. The
finding is in agreement with the study of Sylvia and Etchi, (2019) which emphasized that
emerging technologies such as machine learning can perform effectively well in transaction
classification, analyses and recording into the correct books of account.
In summary, the study reinforces the transformative potential of AI and ML in enhancing the
qualitative attributes, accuracy, and efficiency financial reporting processes, eventually
contributing to improved decision-making and stakeholder trust in financial information.

36
This discussion highlights the key findings of the study in a clear and concise manner,
emphasizing their significance and relevance in the context of financial reporting and emerging
technologies.

37
Recommendations:

Machines can now learn from experience, adapt to new inputs, and carry out activities that
humans would normally be unable to complete thanks to artificial intelligence (AI). The
majority of AI examples you hear about today—from self-driving vehicles to machines
that can play chess—heavily rely on profound learning and the processing of natural
language.
Furthermore, artificial intelligence is here to stay in Pakistan right now! Pakistan is offering
its youth a variety of employment options and sophisticated AI programs. Thus, to argue
that AI has grown to be a crucial component of Pakistan's businesses wouldn't be incorrect.
Whichever industry you work in, artificial intelligence (AI) is present.

Based on the research's findings, the following suggestions are put forth. These suggestions
are meant to tackle the challenges that have been found, enhance the way that things are
done now, and direct further research in this field.

38
Assimilation of Machine Learning Technology (ML):
• Companies should assimilate machine learning technology into their financial
reporting processes and systems. This assimilation or integration will improve data
gathering, recording, classification, analysis, and processing, ultimately ensuring
accuracy and facilitating improved financial decision-making.

Investment in Information Technology Infrastructure:

• Organizations need to invest in Information Technology (IT) architecture capable
of accommodating AI technologies. This investment will provide the necessary
foundation for effective transaction processing and reporting, enabling companies
to leverage AI capabilities efficiently.

Integration of Machine and Deep Learning Algorithms:

• Companies should consider integrating machine and deep learning algorithms into
their accounting software. This integration will empower software systems with
self-learning and autonomous capabilities, allowing organizations to optimize AI
benefits across various financial management processes.

By implementing these recommendations, companies can effectively leverage AI technologies to

enhance their financial reporting processes, leading to increased efficiency, accuracy, and
decision-making capabilities in today's dynamic business landscape. As an enabling foundation
for the introduction of AI technology, organizations should start using Enterprise Resource
Planning (ERP) solutions to promote integration and interplay among all business components and
departments. v. While AI is being used to improve reporting on finances, management should also
make certain that the current workforce, particularly the accounting team, is retrained and upskilled
to use the latest technology.

39
 Conclusion:

Any firm that hopes to stay relevant and competitive in the near future must master the usage
of AI technology. The study's conclusion was that, given the substantial correlation between
the two variables, machine learning's importance in the financial reporting process cannot be
overstated. The study also found that the qualitative aspects of financial reporting are
significantly impacted by artificial intelligence. Effective financial reporting will be facilitated
by setting up data platforms for AI that comprise the necessary instruments for data annotation,
storage, next-generation networks, analytics software, and data protection.
Our lives now revolve around AI and machine learning, both of which will undoubtedly be
relevant in the near future. They improve commonplace technology, revolutionize whole
sectors, spur creativity, resolve challenging issues, and enable customisation.
In this sense, the AI research sector itself must play a crucial role in educating the public about
significant trends and discoveries in a way that is both useful and actionable, devoid of hype
and transparent about the risks and unforeseen repercussions as well as the advantages and
prospects. It is imperative for AI researchers to acknowledge that total independence is not the
ultimate objective for AI systems. As a species, we are strong because we can cooperate to
achieve greater things than any one of us could do on our own. That system for the entire
community must include AI, and there must be open channels of communication between
automated and human decision-makers. Ultimately, the field's success will be determined by
how it has empowered everyone, not by how well machines undercut the same people we are
working to assist.

40
41
References:

• file:///C:/Users/hp/Downloads/EFFECTSOFARTIFICIALINTELLIGENCEONFINAN
CIALREPORTING-EVIDENCEFROMKPMGANDPWC%20(1)%20(1).pdf
• Anne, T. (2005). Pre-attentive processing in vision: Computer vision, graphics, and
image. Retrieved from https://dl.acm.org/doi/10.1016/S0734-189X%2885%2980004-
• Attaran, M. & Deb, P. (2018). Machine Learning: The new 'Big Thing' for competitive
advantage. International Journal of Knowledge Engineering and Data Mining, 5(4),
114-128.
• https://www.spiceworks.com/tech/artificial-intelligence/articles/what-is-ml/
• Chariri, A. (2011). Financial reporting practice as a ritual: Understanding Accounting
within institutional framework. Journal of Economics Business and Accountancy
Ventura, 14(1), 101-118.
• Chris, Brian, Ting & Gary, 2006; Juri & Evangelos, 2019; Mohri, Rostamizadeh &
Talwalkar, 2012; Jarrahi, 2018; John and Darrell, 2020.
• https://affinityvibez.com/blogs/future-of-artificial-
intelligence#:~:text=So%2C%20the%20future%20of%20artificial,role%
20in%20almost%20every%20sector.
• Cindy, G. (2017). Exploring the impact of Artificial Intelligence on the Accounting
profession. Journal of Research in Business, Economics and Management, 3(12),
2395-2210. Clarence G., Gary P., Seow S., Benjamin L. & Melvin Y. (2019). Charting
the future of accountancy with AI. Retrieved from
https://accountancy.smu.edu.sg/news/2019/07/30/charting-future-accountancy-
aismu-accountancy-professorspartner-cpa-australia.
• https://www.spiceworks.com/tech/artificial-intelligence/articles/what-is-ml/
• https://sitn.hms.harvard.edu/flash/2017/history-artificial-intelligence/
• https://kpmg.com/pk/en/home/about/overview/history.html#:~:text=KPMG%20in%20
Pakistan%20embodies%20the,advice%20on%20potential%20business%20risks.
• https://www.spglobal.com/en
• https://www.spglobal.com/en/our-history
42
 • https://sphweb.bumc.bu.edu/otlt/mph-
modules/sb/behavioralchangetheories/behavioralchangetheories4.html
• https://www.sas.com/en_in/insights/analytics/what-is-artificial-
intelligence.html#:~:text=Artificial%20intelligence%20(AI)%20makes%20it,learning
%20and%20natural%20language%20processing.
• https://ai100.stanford.edu/gathering-strength-gathering-storms-one-hundred-year-
study-artificial-intelligence-ai100-2021-
3#:~:text=AI%20needs%20to%20be%20incorporated,we%20are%20trying%20to%
20help

43

View publication stats