Ekiti State University, ado ekiti
Faculty of Multidisciplinary Studies
Department of Mass Communication
Course code: Mac 204
Course Title: Foundations of communication research
Lecturer In charge: Dr. Ajakaiye
Question: As a group, we are asked to pick a research topic (a subject/topic you'd like to make a
research on) and work on 3 things
1. Introduction
2. Literary review and theoretical framework
3. Research methodology
Group Members
Oguntoyinbo Damilare Praise 2303089
Ogunfuye David Oluwatosin 2303095
Kareem Adepeju Balikis 2303092
Shobowale Daniel Ayomide 2303098
Alaketu Sofiat Ifeoluwa 2303090
Boboye Elizabeth Wuraola 2303094
OJO MARTINS EVELYN OLUWATOBINIFE 2303093
Ojikutu Daniel oluwatobiloba 2303091
Kyudu Gifty Joseph 2303097
Yaya Fatima Moyosola 2303099
Ogunlola Oyindamola Adebola 2303096
The Ethical Implications of Artificial Intelligence in Decision-Making
Table of Contents
Title
Dedication
Acknowledgements
Chapter One: Introduction
1.1 Background of the Study
1.2 Statement of the Problem
1.3 Objectives of the Study
1.4 Research Questions
1.5 Significance of the Study
1.6 Scope of the Study
1.7 Definition of Terms
Chapter Two: Literature Review and Theoretical Framework
2.1 Conceptual Review
2.1.1 Artificial Intelligence and Decision-Making
2.1.2 Ethical Concerns in AI Systems
2.1.3 Accountability, Transparency, and Bias in AI
2.2 Empirical Review
2.2.1 Studies on AI Adoption and Public Perception
2.2.2 Ethical Case Studies on AI in Decision-Making
2.3 Theoretical Framework
2.3.1 Diffusion of Innovations Theory
Chapter Three: Research Methodology
3.1 Research Design
3.2 Research Method
3.3 Population of Study
3.4 Sample Size
3.5 Sampling Technique
3.6 Methods of Data Collection
3.7 Methods of Data Analysis
3.8 Reliability and Validity
Chapter Four: Data Presentation and Analysis
4.1 Presentation of Data
4.2 Analysis of Research Questions
4.3 Discussion of Findings
Chapter Five: Summary, Conclusion, and Recommendations
5.1 Summary of the Study
5.2 Conclusion
5.3 Recommendations
5.4 Suggestions for Further Research
References
Appendices
Appendix A: Research Questionnaire
Appendix B: Survey Data Tables
Dedication
This research work is dedicated to the Almighty God, the source of wisdom, knowledge, and
understanding.
We also dedicate it to our families and loved ones, whose unwavering support and encouragement have
been our strength throughout this academic journey.
Acknowledgements
We wish to express our profound gratitude to God Almighty for His guidance, grace, and sustenance
which made the completion of this project possible.
Our heartfelt appreciation goes to the lecturers and academic mentors in the Department of Mass
Communication for their efforts, guidance, and constructive criticism. Their commitment to excellence
and knowledge transfer has been a great inspiration to us.
We also acknowledge the collective contributions of our research group members, whose dedication
and teamwork brought this work to reality:
Oguntoyinbo Damilare Praise
Ogunfuye David Oluwatosin
Kareem Adepeju Balikis
Shobowale Daniel Ayomide
Alaketu Sofiat Ifeoluwa
Boboye Elizabeth Wuraola
Ojo Martins Evelyn Oluwatobinife
Ojikutu Daniel Oluwatobiloba
Kyudu Gifty Joseph
Yaya Fatima Moyosola
Ogunlola Oyindamola Adebola
Finally, we extend our gratitude to our classmates, friends, and everyone who contributed in one way or
another to the success of this research work.
Chapter One: Introduction
1.1 Background of the Study
The 21st century has been marked by rapid technological advancements, among which Artificial
Intelligence (AI) stands out as one of the most transformative. Artificial Intelligence refers to computer
systems or machines that are capable of simulating human intelligence processes such as learning,
reasoning, problem-solving, perception, and decision-making. Unlike earlier technological tools, AI does
not merely assist humans in performing tasks; it increasingly undertakes roles that were traditionally the
exclusive preserve of human judgment.
Today, AI systems are embedded in multiple sectors of human activity. In healthcare, AI systems assist
doctors in diagnosing illnesses, predicting disease outbreaks, and even recommending treatment plans.
In finance, AI algorithms determine loan approvals, detect fraudulent transactions, and manage
investment portfolios. In the justice system, predictive algorithms are deployed in some countries to
recommend bail, probation, or sentencing decisions. Similarly, businesses now use AI to screen job
applicants, recommend products to consumers, and manage supply chains. These applications
demonstrate the widespread reliance on AI in making decisions that directly affect human lives.
Despite these benefits, the rise of AI poses significant ethical questions. Human decision-making, though
imperfect, is shaped by moral reasoning, empathy, cultural values, and social accountability. Machines,
however, operate based on data, algorithms, and programmed logic. If the data provided to AI systems
are biased, incomplete, or discriminatory, the resulting decisions may reproduce existing social
inequalities. For instance, Amazon had to discontinue an AI recruitment tool when it was discovered
that the system discriminated against female applicants because it was trained on historical data that
reflected male-dominated hiring patterns. Similarly, predictive policing tools in the United States have
been criticized for disproportionately targeting Black and minority communities, thereby reinforcing
systemic racial bias.
These examples illustrate how the integration of AI in decision-making can lead to unintended yet
harmful consequences. More importantly, they raise critical questions about fairness, accountability,
and transparency. Who is to blame when an AI system makes an error that affects a person’s life—such
as denying a loan, misdiagnosing a patient, or wrongly implicating someone in a crime? Is it the software
developer, the institution deploying the AI, or the government that regulates technology? Such
questions emphasize the urgent need to study the ethical implications of AI in decision-making.
In Nigeria, the discussion around AI is still emerging. The country has witnessed a gradual adoption of AI
technologies in financial services, agriculture, telecommunications, and e-commerce. For example, many
banks in Nigeria now use AI-driven systems for fraud detection and customer service (e.g., chatbots).
However, there are concerns that these technologies are being adopted without adequate ethical
safeguards or public awareness. Nigeria also faces unique challenges such as weak regulatory
frameworks, digital inequality, and limited enforcement of data protection laws. These challenges make
the ethical implications of AI even more pressing, as vulnerable populations could be disproportionately
affected by biased or unregulated AI systems.
Thus, this study situates itself within this global and national debate, examining the ethical implications
of artificial intelligence in decision-making, with a focus on transparency, accountability, bias, privacy,
and human rights.
1.2 Statement of the Problem
The increasing reliance on AI in decision-making presents both opportunities and challenges. While AI
promises efficiency, accuracy, and cost-effectiveness, it also raises profound ethical dilemmas. Unlike
humans, AI systems lack moral judgment, empathy, and the ability to weigh social values. Their
decisions are determined by algorithms and the datasets on which they are trained. Consequently, when
AI systems are biased or flawed, their decisions can lead to injustice, discrimination, or harm.
The first major problem is bias and discrimination. AI systems trained on biased data may
unintentionally perpetuate inequalities. A recruitment AI system might favor certain demographics over
others, while a loan approval system could deny access to credit to already marginalized groups. This
poses ethical challenges concerning fairness and equality of opportunity.
Secondly, there is the issue of accountability. When AI makes a harmful or unfair decision, it is often
unclear who should be held responsible. Unlike human decision-makers who can be questioned, AI
operates as a “black box,” where its decision-making process is not easily understood. This lack of
transparency raises concerns about trust and accountability in critical decision-making processes.
Thirdly, privacy and data protection represent another area of concern. AI systems rely heavily on
personal data to function effectively. In contexts like healthcare, finance, and governance, sensitive
personal information is collected, analyzed, and stored. Without proper safeguards, individuals’ privacy
may be compromised, and their personal data could be misused.
Finally, in countries like Nigeria, weak regulatory frameworks exacerbate these issues. While developed
nations are already debating AI ethics and passing regulations (such as the European Union’s AI Act),
Nigeria and many African countries are still in the early stages of developing such frameworks. This gap
increases the risk of unethical deployment of AI technologies in ways that may violate human rights,
deepen inequality, and erode public trust in technology.
The central problem this study seeks to address, therefore, is the ethical tension between the benefits
of AI decision-making and the risks it poses to fairness, transparency, accountability, privacy, and human
dignity.
1.3 Objectives of the Study
The main objective of this study is to explore the ethical implications of artificial intelligence in decision-
making. The specific objectives are to:
1. Assess public perceptions and experiences with AI in decision-making.
2. Identify key ethical concerns associated with AI, including bias, accountability, transparency, privacy,
and fairness.
3. Analyze the level of trust and comfort individuals have with AI-driven decisions.
4. Investigate views on responsibility and accountability when AI makes harmful or unfair decisions.
5. Provide recommendations for ethical frameworks and policies to guide AI adoption, particularly
within Nigeria.
1.4 Research Questions
The following research questions guide the study:
1. Have you ever interacted with AI?
2. How comfortable are you with AI being used in everyday decision-making?
3. Do you think AI should be allowed to make decisions such as hiring employees, approving loans, and
giving medical advice?
4. In your opinion, who should be held responsible if an AI system makes harmful or unfair decisions?
5. Do you trust AI to be fair and unbiased in its decisions?
6. What concerns you most about AI making decisions that affect people’s lives?
7. Do you think AI should always explain how it reached a decision (transparency)?
8. How likely are you to accept a decision made by AI over one made by a human expert?
9. Do you believe governments should regulate the use of AI in decision-making?
10. In your own words, what do you think is the biggest ethical issue with AI in decision-making?
1.5 Significance of the Study
This study is significant in several ways.
Academic Contribution: The research contributes to scholarly discourse on AI ethics by providing insights
from the Nigerian perspective, where literature on the subject is still limited. It adds to global debates
on how AI affects fairness, transparency, and accountability in decision-making.
Policy and Regulation: The findings will be useful to Nigerian policymakers, regulators, and legal
practitioners who are beginning to draft policies on AI adoption. By highlighting public concerns and
ethical risks, the study can inform the development of laws and guidelines that ensure responsible AI
use.
Technology Development: The study will also benefit technology developers and companies adopting AI
by emphasizing the importance of ethical considerations in system design and deployment.
Public Awareness: Finally, the research increases awareness among the general public about the
potential risks and benefits of AI in decision-making. It empowers citizens to critically evaluate AI
systems and demand accountability from institutions that deploy them.
1.6 Scope of the Study
This study is focused on the ethical implications of AI in decision-making. While examples from the
global context are considered, the emphasis is on Nigeria, where AI adoption is growing but regulation
remains weak. The study limits itself to decision-making in sensitive areas such as healthcare, finance,
recruitment, and governance, where the ethical consequences are most pronounced. Other areas of AI
application, such as gaming, entertainment, and robotics in manufacturing, are outside the scope of this
research.
1.7 Definition of Terms
Artificial Intelligence (AI): The simulation of human intelligence in machines that are capable of learning,
reasoning, and decision-making.
Decision-Making: The act of choosing a course of action from alternatives; in AI, this refers to algorithm-
driven outcomes that influence human lives.
Ethics: A set of principles that govern what is morally right or wrong in human behavior and
technological application.
Bias: Systematic unfairness in AI decision-making resulting from flawed or discriminatory data inputs.
Transparency: The ability to explain and understand the logic behind AI decision-making processes.
Accountability: The principle that institutions, developers, or policymakers must take responsibility for
the outcomes of AI systems.
Privacy: The right of individuals to control access to their personal data and prevent unauthorized use.
Human Rights: Fundamental rights such as equality, dignity, and fairness that may be upheld or violated
by AI decisions.
Chapter Two: Literature Review and Theoretical Framework
2.1 Conceptual Review
Artificial Intelligence and Decision-Making
Artificial Intelligence (AI) refers to the design of computer systems capable of performing tasks that
normally require human intelligence, including reasoning, problem-solving, perception, and decision-
making (Russell & Norvig, 2021). In decision-making contexts, AI systems process large datasets,
recognize patterns, and generate outcomes or recommendations. Unlike human decision-making, which
incorporates emotions, ethics, and cultural values, AI relies on algorithms and statistical models. This
distinction raises concerns about the fairness and ethical soundness of AI-driven decisions.
Ethics in AI
Ethics is the philosophical study of morality, concerned with principles that define right and wrong.
When applied to AI, ethics questions whether algorithmic decisions respect fairness, equality, privacy,
transparency, and human dignity (Floridi & Cowls, 2019). Ethical challenges in AI decision-making
include:
Bias and Discrimination: AI systems may inherit prejudices embedded in training data, leading to
outcomes that disadvantage women, minorities, or marginalized groups.
Transparency: AI decisions often appear as a “black box,” making it difficult for users to understand how
outcomes were generated.
Accountability: Determining responsibility when AI makes harmful decisions is a major ethical dilemma.
Privacy: The heavy reliance of AI on big data raises concerns about unauthorized access and misuse of
personal information.
AI in Nigeria
The adoption of AI in Nigeria is still at a developing stage, with applications mainly in finance,
agriculture, e-commerce, and telecommunications. Nigerian banks now use AI chatbots, fraud detection
systems, and automated customer support, while agritech startups employ AI to predict yields and
detect crop diseases. However, ethical concerns remain underexplored in the Nigerian context. Weak
regulations, limited digital literacy, and social inequality amplify the risks of biased and unfair outcomes
in AI decision-making (Onyema, 2022).
2.2 Empirical Review
Globally, scholars have examined the ethical implications of AI in decision-making from various
perspectives.
Bias in AI: A study by Buolamwini and Gebru (2018) showed that commercial facial recognition systems
exhibited racial and gender biases, with error rates significantly higher for darker-skinned women
compared to lighter-skinned men. This demonstrates how biased data can reinforce systemic
discrimination.
Transparency and Accountability: Mittelstadt et al. (2016) argued that the lack of explainability in AI
systems undermines accountability. When AI denies a loan or recommends a medical treatment,
individuals affected often cannot challenge or understand the reasoning behind such decisions.
Trust and Public Perception: A Pew Research Center (2022) survey revealed mixed public trust in AI.
While many respondents recognized AI’s efficiency, a large percentage expressed concerns about
fairness, privacy, and loss of human control in decision-making.
AI in Developing Countries: Research by Akinwale (2021) highlighted that in African contexts, the ethical
risks of AI are compounded by weak regulatory systems and inadequate enforcement of data protection
laws. For example, in Nigeria, many citizens are unaware of how their personal data is used by AI-driven
applications, increasing vulnerability to exploitation.
Healthcare Applications: Topol (2019) emphasized that AI in healthcare has improved diagnostic
accuracy, but warned that without ethical safeguards, patients’ rights and data privacy may be
compromised.
Taken together, these studies underscore the urgency of addressing ethical concerns such as bias,
accountability, transparency, and privacy in the deployment of AI systems. However, there remains a
gap in research specifically focusing on Nigeria and the broader African region, where unique socio-
economic and regulatory contexts demand localized analysis.
2.3 Theoretical Framework
This study is anchored on three major ethical and social theories that provide a lens for analyzing AI
decision-making:
1. Utilitarianism
Proposed by philosophers like Jeremy Bentham and John Stuart Mill, utilitarianism emphasizes that
actions are ethical if they produce the greatest good for the greatest number. Applied to AI, this
framework suggests that AI systems should be judged by the extent to which they maximize societal
benefits—such as efficiency, fairness, and improved services—while minimizing harm. However,
utilitarianism has limitations when AI harms minority groups for the sake of majority benefit.
2. Deontological Ethics
Derived from Immanuel Kant, deontology focuses on duties, rights, and moral principles rather than
outcomes. In AI, this perspective argues that decisions must respect fundamental human rights such as
fairness, privacy, and dignity, regardless of efficiency. For example, even if an AI system improves
productivity, it would be unethical if it violates privacy or discriminates against individuals.
3. Socio-Technical Systems Theory
This theory views technology as embedded within social systems, meaning AI cannot be analyzed in
isolation from the social, cultural, and political contexts in which it operates. In Nigeria, for example, AI
adoption must be understood within the framework of weak regulation, social inequality, and limited
public awareness. This theory emphasizes that ethical AI requires not only technical safeguards but also
social accountability and governance structures.
Summary of Literature Review
From the conceptual, empirical, and theoretical insights reviewed, it is clear that while AI holds
significant potential for improving decision-making, its ethical risks are profound. Issues of bias,
transparency, accountability, and privacy dominate global debates, yet local contexts such as Nigeria
remain under-researched. This study therefore builds on global findings while addressing gaps specific to
Nigeria’s regulatory and social environment.
Chapter Three: Research Methodology
3.1 Research Design
The choice of research design is one of the most critical steps in any academic study because it
determines how the research questions will be addressed. This study adopts a survey research design,
which is widely used in social science and educational research to collect data on opinions, perceptions,
attitudes, and behaviors from a relatively large population within a limited period (Creswell, 2014).
The survey design is appropriate for this study for several reasons. First, the nature of the topic, The
Ethical Implications of Artificial Intelligence in Decision-Making, requires the exploration of diverse
human experiences and attitudes toward AI. A survey provides an efficient way of reaching a wide range
of respondents who have varying degrees of interaction with AI technologies. Second, survey research
allows the use of both quantitative and qualitative questions, enabling the researcher to combine
statistical patterns with deeper insights into ethical concerns. Third, given the rapidly evolving nature of
AI, a survey provides a “snapshot” of public perceptions at a particular point in time, which is important
for identifying current ethical debates and concerns.
By adopting the survey design, this study ensures that data collection is systematic, structured, and
focused on answering the specific research questions outlined in Chapter One.
3.2 Research Method
The study employs a mixed-method approach—predominantly quantitative, with a qualitative
component. Quantitative research is concerned with generating data that can be statistically analyzed,
making it useful for measuring the extent to which respondents trust AI, accept AI decisions, or believe
in government regulation of AI (Babbie, 2020). On the other hand, the qualitative aspect of the study
comes from open-ended survey questions, which allow respondents to freely express their thoughts on
the most pressing ethical issues of AI.
This dual approach provides a comprehensive understanding of the subject matter. The quantitative
data offers generalizable insights into trends, while the qualitative responses enrich the findings by
highlighting individual perspectives that may not be captured in statistical summaries.
3.3 Population of the Study
The target population for this study consists of students and young professionals in Nigeria. This group is
particularly relevant for several reasons. Firstly, students and young professionals are among the most
frequent users of AI-powered tools such as chatbots, learning platforms, automated financial
applications, and recommendation systems. Secondly, they represent a technologically active
demographic that is not only exposed to AI but is also likely to be impacted by AI-driven decision-making
in areas such as employment, healthcare, and access to financial services.
The population is estimated to number in the thousands; however, due to constraints of time and
resources, a representative sample will be drawn.
3.4 Sample Size
In survey research, determining the right sample size is essential to ensure that the data collected can
be analyzed meaningfully and is representative of the larger population. For this study, a sample size of
120 respondents is deemed sufficient. This number is large enough to generate statistically relevant data
but still manageable within the timeframe of the research.
The sample size strikes a balance between breadth and depth: it is large enough to provide generalizable
insights into attitudes toward AI, yet small enough to allow for careful administration and analysis of
responses.
3.5 Sampling Technique
The study will utilize a purposive sampling technique. This technique is suitable because not all members
of the general population have meaningful interactions with AI technologies. By deliberately selecting
respondents who have either used AI or are familiar with its applications, the research ensures that the
data collected is relevant to the study objectives.
Purposive sampling has an added advantage in that it increases the likelihood of obtaining thoughtful
responses from individuals who can reflect on ethical implications, rather than from respondents who
have little or no exposure to AI. This makes the sampling process more efficient and aligned with the
research purpose.
3.6 Methods of Data Collection
The primary instrument for data collection will be a structured questionnaire designed by the
researcher. The questionnaire will consist of two sections:
Section A: Demographic Information – This will collect basic background data such as age, gender, level
of education, and occupation. This information will help determine whether demographic factors
influence perceptions of AI.
Section B: Research Questions – This section will directly address the research questions presented in
Chapter One. It will include both close-ended questions (such as multiple-choice and Likert-scale items)
for quantitative analysis and open-ended questions for qualitative analysis.
The questionnaires will be administered in two forms: online (via Google Forms) and hard copies
distributed among students and professionals in selected areas. Using a dual approach increases
accessibility and ensures a higher response rate.
3.7 Methods of Data Analysis
The data collected will be analyzed using both quantitative and qualitative techniques.
Quantitative analysis will involve the use of descriptive statistics such as percentages, frequency
distributions, and charts. This will allow the researcher to present patterns in respondents’ answers
(e.g., the percentage of people who trust AI versus those who do not). Inferential statistics such as chi-
square tests may also be applied to examine relationships between variables such as education level and
attitudes toward AI decision-making.
Qualitative analysis will involve thematic analysis of the open-ended responses. The researcher will
identify recurring themes such as concerns about bias, lack of transparency, or accountability. These
themes will be coded, categorized, and discussed in relation to the literature review presented in
Chapter Two.
By combining these methods, the study ensures that the data analysis is both rigorous and
comprehensive.
3.8 Reliability and Validity
To ensure the quality of the research instrument, both reliability and validity will be carefully
considered.
Reliability refers to the consistency of the research instrument in measuring what it is intended to
measure. To establish reliability, a pilot study will be conducted with 10 respondents before the main
data collection. The pilot will help identify ambiguous or misleading questions, and adjustments will be
made accordingly.
Validity ensures that the questionnaire accurately measures the ethical implications of AI in decision-
making. To ensure content validity, the questionnaire will be reviewed by two experts in ethics and
technology from the university’s faculty. Their feedback will confirm whether the items align with the
objectives of the study. Additionally, construct validity will be ensured by designing questions that
directly reflect ethical themes such as bias, transparency, accountability, and fairness.