CHAPTER TWO

LITERATURE REVIEW

2.0 Introduction

This chapter will review the related literature on the topic of the study. The review firstly
looks at the conceptualisation of the key elements, review of empirical literature or related
studies, review of the theoretical framework, and justification. The conceptualisation will
bring meanings to the key elements as they shall be used in the study, while review of
empirical literature shows what others have documented on related studies and the theoretical
framework speaks to the theories that guide the study. During the literature review, the views
of varying academics are discussed and compared. The literature review finally serves to
reveal the gaps that the current research is expected to address.

2.1 Conceptualizing

2.1.1 Artificial intelligence (AI)

Artificial intelligence (AI) is the simulation of human intelligence processes by machines,

particularly computer systems (Russel & Norvig, 2021). From this definition it can be
deduced that the referred processes include learning, which involves the acquisition of
information and the associated rules for using it; the reasoning aspect, which denotes using
the rules to reach conclusions; and self-correction. McCarthy (2020) defined AI as the science
and engineering of making intelligent machines, especially intelligent computer programs.
Kaplan and Haenlein (2019) defined AI as a system’s ability to correctly interpret external
data, to learn from such data, and to use those learnings to achieve specific goals and tasks
through flexible adaptation. The aforementioned definitions of AI emphasise both the
scientific and engineering aspects of AI. They highlight the interaction between an AI system
and their environments, focusing on the decision-making and action-taking capabilities of
intelligent agents. Nilsson (2018), in his work, defined AI as the design of intelligent agents
that can perform tasks that typically require human intelligence. This definition encompasses
a broad range of capabilities, including reasoning, learning, and problem-solving, which are
essential for mimicking human-like intelligence.

The rise of generative AI, particularly through large language models, has significantly
impacted various sectors, including education, research, and industry, by enhancing
productivity and creativity (Weaver, 2024). However, the integration of AI into these domains
raises important considerations regarding transparency, ethical use, and the need for proper
citation practices. As AI tools become more prevalent, the traditional citation methods may
not adequately capture the unique contributions of AI in the research process, prompting the
development of frameworks like the Artificial Intelligence Disclosure (AID) Framework,
which aims to provide clarity on how AI tools are utilised throughout various stages of
research and writing (Weaver, 2024). This evolving landscape necessitates ongoing
discussions about the implications of AI on professional integrity and the nature of authorship
in scholarly work (Brown University, 2023; Dalhousie University, 2023).

Although artificial intelligence (AI) has garnered interest for several decades, a universally
accepted definition remains elusive in literature. This absence of a clear definition hampers
empirical studies and complicates a comprehensive understanding of AI in its entirety. To
construct a coherent framework for understanding AI, it is essential first to examine the
concept of "intelligence" before applying it to machines and defining "artificial intelligence."
To effectively measure the intelligence of various technologies categorised as AI, we must
step back from specific systems and establish the foundational elements of what we aim to
capture with the term "intelligence." Legg and Hutter (2020) propose an integrated definition
of intelligence, describing it as "the ability to interact, learn, adapt, and utilise information
from experiences, as well as to deal with uncertainty." Furthermore, the term "artificial"
conveys the idea of human-made objects that mimic or replicate something found in nature.
By synthesising these two fundamental concepts, it becomes imperative to cultivate a more
complex understanding of AI (Legg & Hutter, 2020).

All definitions of artificial intelligence (AI) focus on the replication or enactment of human-
like behaviours by machines. A common theme across these definitions is the endeavour of
AI to mimic human cognitive processes to address various situations. Each definition
emphasises AI's role in emulating human learning mechanisms, processing information, and
tackling problem-solving scenarios. However, a notable exception is found in the definition
by Poole & Mackworth (2022), who discuss the properties of agents without attributing any
human-like characteristics to them. Drawing from these definitions and the clarification of
the two constituent terms that form the overall concept, this research proposes an integrative
definition of AI that will be employed throughout this study. Therefore, this study defines AI
as follows: "the ability of a system to identify, interpret, make inferences, and learn from data
to achieve predetermined organisational and societal goals."
2.1.2 Recruitment and Selection

Recruitment is defined as “the process of identifying and attracting potential candidates for
an organisation to fill open positions” (Armstrong & Taylor, 2020). Dessler (2020) posits that
selection is “the process of choosing the most suitable candidate from a pool of applicants to
fill a position within an organisation.” Some of the methodologies that are deployed in
executing selection include interviews, assessments, and psychometric tests to evaluate a
candidate’s qualifications and suitability. From the recruitment definition, a key theme
emerges, that of a process that involves activities that include advertising job openings,
sourcing candidates, and building a talent pipeline. According to Dessler (2000), recruitment
incorporates activities such as job postings, advertising, and networking to find candidates,
whereas selection involves evaluating candidates through interviews, tests, and background
checks to determine the best fit for the role. The two functions, recruitment and selection, are
critical components of human resources management, a field that encompasses a wide range
of activities aimed at recruiting, developing, and retaining employees to enhance
organisational performance and achieve business objectives. Recruitment and selection are
crucial as they determine the quality of talent that enters the organisation (Bratton and Gold,
2017). Effective HRM practices ensure that the right individuals are hired, which can lead to
improved productivity and reduced turnover (Kaufman, 2021).

With the advent of AI technologies, recruitment and selection have become more and more
automated, raising concerns about ethical implications. Gupta and Shaw (2023) posit that AI-
driven recruitment tools now use predictive analytics to match candidates to job descriptions,
conduct video interviews with facial recognition analysis, and automate resume screening.
While it is believed that AI-driven recruitment systems enhance efficiency, concerns have
been raised with regards to fairness, discrimination, and privacy.

Over the past decade, the integration of artificial intelligence (AI) into recruitment and
selection processes has rapidly increased. Going forward, AI promises to streamline the
hiring process by efficiently processing large volumes of applicant data. Despite this
advantage, AI also raises significant ethical concerns. There are issues such as algorithmic
bias, transparency, accountability, and fairness that have emerged as critical areas that require
further exploration by researchers and practitioners in human resources management (Bogen
& Rieke, 2018; Barocas & Selbst, 2016). This research study therefore seeks to explore these
ethical dimensions in light of recruitment and selection, drawing upon a range of scholarly
works to highlight the challenges and potential strategies to mitigate and reduce negative
outcomes.

2.2 Effects of AI adoption on recruitment and selection

AI adoption and usage are intended to make HR practices easier and faster, with potential
benefits including automation of complex and repetitive tasks, enhanced decision-making and
personalisation, and improved customer experiences. Kissinger, Schmidt, and Huttenlocher
(2021) add an explanation to one of the benefits, outlining that AI systems can take over
tedious, high-volume tasks that are difficult or impractical for humans, improving efficiency
and productivity. In recruitment and selection, AI has changed processes through automation
of repetitive tasks, improving decision-making, and reducing hiring biases (Black &
Johnston, 2022). Artificial intelligence is revolutionising various aspects of Human Resource
practices by streamlining processes, enhancing efficiency, and facilitating data-driven
decision-making. Bersin (2022). The several aspects of AI’s impact on recruitment and
selection are explored hereunder.

2.2.1 Automation and Streamlining of Recruitment and Selection Processes

Incorporating AI-powered solutions plays a pivotal role in automating diverse HR tasks and
processes, presenting an opportunity for organisations to refine and augment their human
capital management efforts. AI is being used to automate hiring tasks such as job postings,
resume screening, candidate ranking, and interview scheduling. According to Smith and
Brown (2021), AI-driven tools can reduce recruitment time by 30-50%. The researchers posit
that this allows HR practitioners to focus on strategic functions. In addition, AI-powered
applicant tracking systems (ATS) streamline candidate shortlisting through the scanning of
curriculum vitae/resumes for the relevant keywords, skills, and experience to which the AI
tools are trained on. It can be therefore deduced that AI facilitates the automation of tedious
and repetitive tasks inherent in recruitment and selection, including resume screening and
applicant tracking. Through sophisticated algorithms, AI systems swiftly process vast
amounts of data, discern patterns, and make informed decisions based on objective criteria,
thereby liberating HR professionals to concentrate on strategic endeavours with higher value.
Balasundaram and Venkatagiri (2020) cited a McKinsey study indicating that modern
technologies have the capability to automate around 56 percent of the conventional 'hire to
retire' tasks, with the flexibility to accommodate unlimited process modifications.
AI algorithms excel in swiftly analysing extensive datasets, recognising patterns, and making
informed decisions based on data insights, outperforming conventional manual methods. An
example of the recruitment and talent acquisition process by Bersin (2022) is that AI-powered
tools are being employed for resume screening, where they analyse resumes and job
applications to pinpoint suitable candidates based on specific keywords, skills, and
experience. These tools also excel in candidate matching by employing sophisticated
algorithms to pair candidates with job openings that align with their qualifications and
preferences. Additionally, AI-driven chatbots are becoming increasingly popular in
interacting with job applicants, addressing their queries, scheduling interviews, and offering
feedback, thereby elevating the overall candidate experience. Geiger (2020) suggests that
standardising processes aims to optimise productivity by maximising output while
minimising input, ultimately enhancing organisational performance through streamlined
operations.

2.2.2 Enhanced Candidate Experience

Through AI chatbots, AI improves candidate engagement in the process of recruitment as AI-

powered chatbots provide real-time feedback, answer any frequently asked questions (FAQs),
and guide candidates throughout the job application process (Lee & Kim, 2023). Adam et al.
(2024) made a finding that organisations that use AI in recruitment reported increased
candidate satisfaction due to faster response times and personalised experiences. These
assertions are further confirmed by Korn Ferry (2024), where they reported that AI augments
the candidate experience by providing timely updates and feedback, which can improve the
overall perception of the recruitment and selection process.

2.2.3 Facilitates data-driven decision making

AI-driven recruitment tools are all designed to leverage Big Data Analytics (BDA) and
Machine Learning (ML) to predict candidate success based on predetermined historical
organisational hiring data (Li & Zhao, 2023). This augments the accuracy of the selection
process and minimises reliance on subjective assessments. By harnessing AI-driven tools in
recruitment and selection, organisations benefit from AI’s analytical capabilities, fully
utilising their data to drive decision-making processes effectively. AI systems excel at swiftly
analysing vast and intricate datasets, uncovering patterns and insights that may elude human
observation, and offering data-driven recommendations and predictions. Babu et al. (2024)
posit that this data-centric methodology, facilitated by AI, enables organisations to depart
from traditional reliance on intuition or past experiences and embrace evidence-based
decision-making.

Despite the mentioned implications of AI on recruitment and selection, ethical concerns arise
regarding data privacy and bias that is inherently embedded in AI models. While AI increases
efficiency, it also raises concerns regarding fairness and transparency. This is so because
algorithms' biases present in historical hiring data may unintentionally reinforce biases that
are present in historical hiring data (Binns, 2018). While AI has the potential to reduce the
inherent human bias in recruitment and selection, it can also perpetuate existing biases if not
managed and deployed properly. However, studies have revealed that AI can help standardise
the evaluation process, leading to more objective hiring decisions (Mori et al., 2024).

2.3 Challenges associated with AI in recruitment and selection.

The integration of AI in recruitment and selection processes has raised ethical concerns,
incorporating algorithmic bias, lack of transparency, and data privacy issues. Unconsciously,
AI systems can perpetuate existing biases present in the training data, leading to unfair
treatment of certain groups of candidates in the recruitment and selection process (Forbes,
2023). Below is an exploration of the ethical implications of AI in recruitment and selection.

2.3.1 Algorithmic Bias and Fairness

Binns (2018) explored fairness in machine learning with specific emphasis on recruitment. A
finding was made that AI systems can reinforce existing biases if they are trained on biased
data. For example, if historical hiring data reflects gender or racial biases, the AI system may
continue to favour certain groups over others (Hunkenschroer & Luetge, 2022). Binns (2018)
saw that hiring algorithms tend to favour candidates from certain demographics, hence
perpetuating inequality. To buttress this position, another study by Crawford (2017) focused
on how AI systems reflect the biases that are present in the datasets they are trained on. It was
Crawford’s observation that this can result in discriminatory practices in recruitment,
selection, and promotions. Crawford posits that this often happens without human resources
practitioners being aware of it.

Research by Dastin (2018) highlights a case where Amazon developed a recruitment tool that
favoured male candidates over female candidates, as it was based on resumes submitted to
the company over a decade, which were predominantly male. Such biases challenge the
ethical principle of fairness and equal opportunity, which are fundamental in recruitment and
selection processes.

2.3.2 Transparency and Accountability

Kim (2017) posits that without clear explanation on how algorithms make decisions, it
becomes difficult to ensure accountability and fairness in the recruitment process. This
narrative is further buttressed by Pasquale (2015) in what has been termed the “Black Box
Society,” where he discussed the concept of black box algorithms, where decision-making
processes are not transparent. As a result, this opacity becomes problematic in recruitment
and selection as employees and applicants may not understand why certain decisions are
made, for example, hiring or rejection decisions. This leads to reduced trust in AI-driven
recruitment and selection processes

The ethical principle of transparency is crucial in the recruitment process. Candidates should
be informed about how AI systems evaluate their applications. Hwang et al. (2020)
emphasise that organisations must disclose the algorithms and criteria used in AI recruitment
processes to foster trust among candidates. Furthermore, accountability mechanisms must be
established to ensure that organisations are held responsible for the outcomes produced by AI
systems, particularly in cases where biases lead to discrimination.

Informed consent is essential in ethical recruitment practices. Candidates should be aware of

how their data will be used and the criteria for evaluation. Research indicates that many
organisations do not adequately inform candidates about the use of AI in recruitment, which
raises ethical concerns regarding autonomy and informed decision-making (Smith, 2021).
Organisations must ensure that candidates consent to the use of their data in AI recruitment
processes.

2.3.3 Data Privacy and Security

As the integration of AI systems with the recruitment and selection processes continues, AI
systems are collecting vast amounts of candidate data, raising concerns over data storage,
access, and security (Zuboff, 2019). Cappelli et al. (2019) posit that during the process of
recruitment and selection, many candidates are unaware of how their data is being used,
stored, or shared, raising ethical questions about consent and data security. As organisations
increasingly rely on AI to process applicants’ data, adherence to data protection regulations
becomes paramount. Studies indicate that many organisations struggle to comply with
regulations, such as the General Data Protection Regulation (GDPR)—a European Union law
that regulates how organisations collect, process, and store personal data of individuals within
the EU, risking legal repercussions and ethical breaches (Smith & Wesson, 2020).
Organisations must implement robust data governance frameworks to ensure compliance and
protect applicants’ data. The ethical implications of data ownership are significant as
applicants lack clarity regarding who owns the data generated through AI systems and how it
will be used. Research suggests that organisations should establish clear policies regarding
data ownership and usage to uphold ethical standards and promote transparency (O'Neil,
2016). This clarity helps mitigate ethical concerns related to data exploitation.

The security of applicants’ data is another critical ethical consideration. Organisations must
implement stringent security measures to protect sensitive applicants’ information from
breaches or unauthorised access. Research indicates that organisations that fail to secure
employee data face not only legal consequences but also reputational damage (Smith &
Wesson, 2020). Ethical data practices require organisations to prioritise data security as a
fundamental responsibility.

2.3.4 Governance frameworks used to organisations against ethical challenges

associated with AI driven recruitment and selection

The governance of AI in recruitment and selection is guided by several key frameworks

aimed at ensuring ethical practices. At the global level, the European Union AI Act
establishes a risk-based classification for AI systems, mandating transparency, human
oversight, data quality, and accountability in recruitment processes (European Commission,
n.d.). This Act requires organisations to inform candidates about AI involvement and ensures
that human intervention is available to address potential biases in decision-making.

Complementing this is the General Data Protection Regulation (GDPR), which focuses on
data privacy and fairness by stipulating that personal data must be processed lawfully and
transparently (European Parliament & Council, 2016). It grants candidates the right to
explanations regarding automated decisions and limits data collection to what is necessary for
hiring. Additionally, the OECD AI Principles advocate for fairness, transparency, and
accountability, urging organisations to monitor AI systems for bias and maintain human
oversight in final hiring decisions (OECD, 2020). Together, these frameworks provide a
robust foundation for ethical AI use in recruitment, particularly relevant to the Zimbabwean
context.
At the regional level, African countries have now begun developing AI strategies to ensure
responsible and ethical AI use. In the South African jurisdiction, for example, the Presidential
Commission on the Fourth Industrial Revolution proposed frameworks for ethical AI use
(Hassan, 2020). However, the research gaps this exploration seeks to uncover are evident, as
these frameworks often lack enforceability and specificity in recruitment and selection
processes (Hassan, 2020). In addition, the African Union (AU) Convention on Cyber Security
and Personal Data Protection (2014) makes provisions for AI and data protection guidelines
(Kanu, 2023). However, the convention lacks enforcement mechanisms, as countries have not
yet domesticated its provisions into their local contexts, and at the AU level, this has not been
made mandatory (Kanu, 2023). The African Union (AU) Executive Council endorsed the
Continental Artificial Intelligence strategy in 2024. Despite lacking specificity in recruitment
and selection, this shows Africa’s commitment to an Africa-centric development focused
approach to AI (African Union, 2024).

In Zimbabwe, the integration of Artificial Intelligence (AI) technologies within organisational

frameworks has become increasingly prevalent, paralleling global trends. However, the
ethical challenges associated with AI implementation pose significant risks to organisations,
including issues related to bias, privacy, accountability, and transparency (Chikowore &
Chigona, 2021). As AI systems are adopted across various sectors, from agriculture to
healthcare, it is paramount for organisations to develop effective governance strategies that
not only comply with regulatory frameworks but also align with ethical standards. This
review seeks to assess the governance strategies currently employed by organisations in
Zimbabwe to mitigate the ethical challenges arising from AI technologies on recruitment and
selection, if any. It will explore the effectiveness of these strategies, the existing gaps in
governance, and provide recommendations for enhancing ethical AI practices.

The governance frameworks guiding the use of artificial intelligence (AI) in Zimbabwe are
still in their infancy, reflecting a broader trend across many developing nations where
regulatory structures have not yet fully adapted to the rapid advancements in AI technology.
As of 2024, Zimbabwe has made strides in ICT regulation through the National ICT Policy,
which was launched in March 2024; however, there remains a significant gap in specific AI
governance policies that address the unique challenges and opportunities presented by AI
technologies (Hlongwane et al., 2024). The lack of a comprehensive AI policy framework in
Zimbabwe raises concerns about ethical considerations, data privacy, and the potential for
algorithmic bias, which are critical issues that need to be addressed to foster trust and
accountability in AI applications (Du, 2024c). The ethical implications of AI usage are
particularly pressing in the context of recruitment and selection, where biases in AI
algorithms can lead to discriminatory practices in hiring and employee evaluation processes
(Binns, 2018). Furthermore, the absence of clear guidelines on data protection exacerbates
these challenges, as organisations may struggle to comply with existing laws while
implementing AI systems that require extensive data collection and processing (European
Commission, 2016).

In the context of labour legislation, recruitment and selection in Zimbabwe are primarily
governed by the Labour Act, Chapter 28:01, and related subsidiary legislation that are
industry-specific to some extent. The Labour Act provisions ensure fairness and prohibit
discrimination in hiring practices (Labour Act, Cap 28:01). Despite the Labour Act not
mentioning AI, its anti-discrimination and equality provisions apply to decision-making
processes, including those that are augmented by AI.

In recent years, Zimbabwe has begun recognising the importance of ethical governance in AI.
The Cyber and Data Protection Act (Chapter 11:12) of 2021 establishes guidelines for the
collection and processing of personal data. This legislative effort reflects a growing
awareness of the need for regulatory frameworks that address privacy concerns, which are
particularly relevant in the context of AI (Mavhunga, 2020). Section 14 of the Cyber and
Data Protection Act outlines the rights of data subjects, emphasising the requirement for
explicit consent before processing personal data. This therefore means AI systems used in
recruitment must ensure candidates provide this consent. In addition, section 10 details the
conditions under which non-sensitive data can be processed, such as obtaining consent from
the data subject or a competent person. Furthermore, section 11 discusses the processing of
sensitive information, requiring written consent from candidates when such data is involved.
This overarching provision in the Cyber and Data Protection Act impliedly speaks to issues
around consent and processing of data in recruitment and selection.

In terms of quality and security of data, Section 7 of the Cyber and Data Protection Act
establishes the Quality of Data principle, mandating that data processed must be accurate,
relevant, and not excessive for its intended purpose. AI tools must ensure they do not rely on
outdated or incorrect data. Section 18 of the same Act requires data controllers, which are
defined in the Act as a person or a company that determines the purpose and means of
processing personal data either jointly or alone, to implement appropriate technical and
organisational measures to protect personal data from unauthorised access or breaches.
Section 14 reiterates the rights of data subjects, including access to their personal data, the
right to request corrections, and the right to object to automated decision-making processes,
which are particularly relevant in AI recruitment.

On transparency and accountability section 23 of the Cyber and Data Protection Act imposes
a requirement for openness in processing, ensuring candidates are informed about how their
data will be used. This by implication includes the use of AI in recruitment decisions. Section
24 emphasizes accountability, requiring employers to demonstrate compliance with the Act.
This therefore includes documenting AI processes and their adherence to data protection
principles. The Cyber and Data Protection Act, in section 12, addresses the processing of
genetic, biometric, and health data, outlining strict conditions and the necessity for consent
when such sensitive data is processed. All this data is used in recruitment and selection as
organisations try to settle for the best candidate using defined criteria. From the aforesaid,
Cimas Health Group, through this Act, has become a data controller, and as part of
compliance, a data protection officer has undergone a data protection certification course with
the Postal and Telecommunications Authority of Zimbabwe (POTRAZ), the government
entity that is now designated as the country’s Data Protection Authority (DPA).

Moreover, organisations are increasingly implementing internal governance frameworks that

align with ethical standards for AI. These frameworks typically involve the establishment of
ethical guidelines, policies, and procedures to govern AI development and deployment. For
example, some organisations have adopted ethical AI principles that emphasise fairness,
accountability, and transparency in their AI systems (Munyoro, 2021). Such principles aim to
mitigate bias by ensuring that AI algorithms are trained on diverse and representative
datasets. However, the practical application of these principles often faces challenges,
particularly in resource-constrained environments where organisations may lack the
necessary technological infrastructure or expertise to implement ethical AI practices
effectively (Chikowore & Chigona, 2021).

The research study case Cimas Health Group (Private) Limited is ISO 27001 certified. ISO
27001 is an international standard that provides a framework for organisations to establish,
implement, maintain, and continually improve an Information Security Management System
(ISMS) to protect their information assets. It focuses on identifying and managing
information security risks, ensuring confidentiality, integrity, and availability of data.
To Cimas Health Group, ISO 27001 requires the organisation to establish, implement, and
maintain an ISMS, which ensures that sensitive data is handled securely. The ISMS requires
regular risk assessments to help identify potential ethical issues related to data misuse or bias.
By implication, bias includes those in AI algorithms in recruitment and selection. The ISMS
also provides for confidentiality and integrity of data. It provides that personal data collected
is protected against unauthorised access and breaches, aligning with ethical standards of
candidate privacy. This aligns with recruitment and selection. The ISMS also makes a critical
provision on data minimisation, encouraging the collection of only necessary data, reducing
the risk of discrimination or bias in AI-driven decisions.

Governance frameworks guiding the use of AI in Zimbabwe are still developing, with
significant work needed to create comprehensive policies that address ethical, legal, and
social implications. The Data Protection Act (2021) governs data privacy but does not
specifically address AI or its ethical use in recruitment and selection. As previously alluded
to, this lack of specificity then creates challenges in ensuring fairness and transparency in AI-
driven recruitment and selection. Comparing Zimbabwe to some global frameworks such as
the EU’s GDPR and the OECD principles reveals a glaring fact: the global frameworks
provide more comprehensive guidance on ethical AI use. As such, Zimbabwe could benefit
from adopting some principles as we domesticate AI governance through our own policies
and frameworks. The establishment of a clear regulatory framework, coupled with
stakeholder engagement and alignment with international standards, will be crucial in
navigating the complexities of AI governance in Zimbabwe.

2.3.5 Effectiveness of existing governance frameworks in mitigating ethical challenges.

The rapid advancement of Artificial Intelligence (AI) technologies has had a profound impact
on various sectors globally, including healthcare, finance, and education. In Zimbabwe,
where the adoption of AI is still in its infancy, the potential ethical challenges associated with
its implementation are significant and require urgent attention. These challenges encompass
issues such as bias, privacy, accountability, and transparency, which can adversely affect
individuals and society at large as discussed previously, aligned to recruitment and selection.
This section aims to assess the effectiveness of existing governance frameworks in Zimbabwe
in mitigating these ethical challenges associated with AI in recruitment and selection. By
examining the current regulatory landscape, institutional frameworks, and practical
applications, this review will provide insights into the strengths and weaknesses of
Zimbabwe’s approach to AI governance in recruitment and selection and in general.

While the Zimbabwean Cyber and Data Protection Act regulates personal data protection, it
lacks explicit AI ethics guidelines (Chihambakwe & Nyasha, 2024). This therefore means this
legislation has a gap, as there is no specific alignment to recruitment and selection. However,
while this act is a positive step, its effectiveness largely depends on its implementation and
the extent to which organisations adhere to its provisions. Currently, many organisations lack
a comprehensive understanding of data governance, leading to potential ethical breaches in
AI applications (Chikowore, 2022). These AI applications include those deployed in
recruitment and selection.

The effectiveness of governance strategies in safeguarding organisations against ethical

challenges associated with AI in Zimbabwe is further influenced by the broader socio-
political context. The regulatory environment in Zimbabwe is characterised by political
instability and limited resources, which can impede the implementation of effective
governance frameworks (Munyoro, 2021). Organisations may face challenges in navigating
bureaucratic hurdles when attempting to comply with regulatory requirements. Furthermore,
the lack of a cohesive national strategy for AI governance hinders efforts to establish a
unified approach to addressing ethical challenges (Mavhunga, 2020).

From a surface reading and assessment, ISO 27001 is a sound and effective governance
framework for mitigating ethical implications associated with AI in recruitment and selection.
By emphasising risk management, data protection, clear policies, stakeholder engagement,
monitoring, and accountability, the standard helps Cimas Health Group to navigate the
complex ethical landscape of AI technologies. However, due to the rapid development of AI
technologies, it is essential to complement ISO 27001 with additional ethical frameworks and
guidelines specifically focused on AI to address the unique challenges posed by these
technologies (ISO, 2023).

ISO/IEC JTC 1/SC 42 is a specific framework from ISO that focuses on artificial intelligence
(AI) and complements ISO 27001. This subcommittee develops international standards
related to AI, addressing various aspects, including governance, risk management, and ethical
considerations.

On its own, ISO 27001 is not effective in addressing the ethical implications of AI in
recruitment and selection. ISO provides key standards and guidelines from the subcommittee
ISO/IEC JTC 1/SC 42 ISO/IEC 42001:2023 - AI Management System is a standard that
provides a framework for establishing, implementing, and maintaining an AI management
system, which Cimas is not yet certified on. It emphasizes governance, ethical considerations,
and compliance, which are crucial for organizations using AI technologies (ISO, 2023). In
addition, there is ISO/IEC TR 24028:2020 - Overview of Trustworthiness in AI, which is a
technical report that outlines the key concepts of trustworthiness in AI systems, including
transparency, accountability, and ethical use. It serves as a guide for organisations to ensure
that their AI implementations are reliable and aligned with ethical standards (ISO, 2020).
Furthermore, there is ISO/IEC TR 24029-1:2021 - Assessment of AI Systems, another
technical report that provides guidelines for assessing the trustworthiness of AI systems,
focusing on aspects such as bias, fairness, and performance. It helps organisations evaluate
the ethical implications of their AI technologies (ISO, 2021). ISO/IEC 23894:2021 - Ethics in
AI is another standard that focuses on ethical considerations in AI development and
deployment, addressing issues such as fairness, accountability, and transparency. It provides a
framework for organisations to integrate ethical principles into their AI practices (ISO, 2021).

To address these challenges, several recommendations have been made to enhance

governance strategies for AI in Zimbabwe. First, organisations should prioritise the
development of comprehensive AI governance frameworks that explicitly address ethical
considerations. This involves not only establishing ethical guidelines but also implementing
robust monitoring and evaluation mechanisms to assess the effectiveness of these frameworks
(Chikowore, 2022). Furthermore, investing in employee training and capacity building is
essential for fostering a culture of ethical awareness. Organisations should develop tailored
training programs that equip employees with the knowledge and skills necessary to navigate
the ethical complexities of AI technologies.

Additionally, promoting multi-stakeholder engagement is crucial for developing inclusive

governance frameworks. Organisations should actively collaborate with government
agencies, civil society, and academia to create platforms for dialogue and knowledge-sharing
(Munyoro, 2021). These collaborations can help identify and address ethical challenges more
effectively while ensuring that diverse perspectives are considered in decision-making
processes. Furthermore, organisations should establish clear accountability mechanisms that
delineate roles and responsibilities in AI governance. This can enhance transparency and
build trust among stakeholders, ultimately contributing to more responsible AI practices
(Chikowore & Chigona, 2021). If these are put in place, recruitment and selection as a
process will benefit within the HR field.

As the adoption of AI technologies grows, so do the ethical implications associated with their
use. In Zimbabwe, the integration of AI is still emerging, with notable developments in
sectors such as agriculture, healthcare, and financial services. However, without a coherent
national strategy for AI adoption, ethical governance becomes increasingly complex and
challenging (Chikowore & Chigona, 2021). Ethical challenges such as algorithmic bias,
which occurs when AI systems produce outcomes that are prejudiced due to flawed training
data, can lead to discriminatory practices in recruitment and selection. Furthermore, privacy
concerns arise from the extensive data collection required for training AI systems. The lack of
robust data protection laws in Zimbabwe exacerbates these issues, as individuals may be
unaware of how their data is being used and the potential risks associated with its collection
(Chigona & Chikowore, 2021). These sentiments equally apply to the process of recruitment
and selection, where vast amounts of data are collected during these processes.

The ethical implications of AI in Zimbabwe are further complicated by accountability issues.

When AI systems make decisions that negatively impact individuals, determining who is
responsible can be ambiguous. This is particularly concerning in sectors such as healthcare
and criminal justice, where AI technologies may influence critical outcomes affecting
people's lives (Chikowore, 2022). The "black box" nature of many AI systems, which
obscures the decision-making processes, contributes to the challenges of accountability and
transparency. Stakeholders, including policymakers and the public, may lack a clear
understanding of how AI systems operate, making it difficult to trust their outcomes
(Munyoro, 2021).

The existing governance frameworks in Zimbabwe, while showing some potential, are not yet
adequately equipped to address these multifaceted ethical challenges of AI in recruitment and
selection. The Data Protection Act of 2021 aims to establish guidelines for the collection,
storage, and processing of personal data. While this act represents a step toward addressing
privacy concerns, it fails to specifically address the ethical implications associated with AI
technologies in recruitment and selection specifically. The Information and Communication
Technology (ICT) Policy emphasises the importance of technology in national development
but lacks specific provisions for AI governance (Mutimukulu, 2021). This lack of specificity
is a critical gap in the current governance landscape, as it leaves many ethical challenges
unaddressed, inclusive of those required in recruitment and selection.

Institutionally, the Zimbabwean Ministry of ICT, Postal and Courier Services oversees the
adoption of technology, including AI. However, its focus has primarily been on infrastructure
development rather than ethical governance. The ministry’s efforts are hampered by limited
resources and a lack of dedicated personnel to address AI-related ethical issues (Mavhunga,
2020). Furthermore, the Zimbabwean National Commission for UNESCO has initiated
programs to promote ethical standards in technology use but has limited influence on AI
governance. Its focus on education and cultural preservation does not sufficiently address the
urgent need for comprehensive AI governance frameworks (Chigona & Chikowore, 2021).

International frameworks and guidelines, such as the OECD Principles on Artificial

Intelligence and the European Union's AI Act, provide valuable benchmarks for Zimbabwe to
consider. These frameworks emphasise the importance of ethical considerations,
transparency, accountability, and stakeholder engagement in AI governance (OECD, 2021).
However, Zimbabwe has yet to fully align its governance frameworks with these international
standards, which could enhance its capacity to mitigate ethical challenges associated with AI
in recruitment and selection.

Assessing the effectiveness of existing governance frameworks in Zimbabwe reveals both

strengths and weaknesses. One strength is the growing commitment to data protection as
reflected in the Cyber and Data Protection Act. This indicates a recognition of the importance
of safeguarding personal data and building trust in AI technologies (Chikowore & Chigona,
2021). Additionally, the ICT Policy acknowledges the transformative potential of technology
in national development, laying a foundational step toward fostering an environment
conducive to ethical AI governance. However, the weaknesses of existing frameworks are
pronounced. The lack of specificity regarding AI technologies and their ethical implications
leaves critical gaps in governance. Moreover, there is insufficient stakeholder engagement in
the development of AI governance frameworks. The limited involvement of civil society,
academia, and the private sector in discussions around AI governance can result in
regulations that do not fully address the concerns of all stakeholders (Munyoro, 2021). In
October 2024, the government announced that the National AI Policy framework was now
complete (The Herald, 14 October 2024). However, much focus, as per the announcement,
was more on embracing digital technologies without addressing the real governance issues
that other jurisdictions have focused on in terms of AI governance.

Implementation challenges further complicate the governance landscape. Even when

regulations exist, their effectiveness is often hindered by limited resources, inadequate
training, and a lack of awareness about AI technologies among policymakers and the public.
For instance, a study by Chikowore (2022) highlighted that many public officials lack a basic
understanding of AI technologies, which hampers their ability to develop informed policies.
Additionally, the absence of dedicated regulatory bodies for AI governance limits oversight
and accountability. Existing institutions may not have the capacity or expertise to monitor AI
implementations effectively, leading to potential ethical breaches (Mutimukulu, 2021).

To improve the governance frameworks for AI in Zimbabwe, several recommendations can

be made. First, the government should develop comprehensive AI policies that specifically
address ethical considerations. This includes defining ethical standards, accountability
mechanisms, and transparency requirements for AI systems. Engaging stakeholders from
various sectors, including civil society, academia, and the private sector, is crucial for
developing effective governance frameworks. Collaborative efforts can lead to more inclusive
regulations that address the concerns of all stakeholders. Furthermore, investing in capacity
building and education is vital for effective governance. Training programs for policymakers,
technologists, and the public can enhance understanding of AI technologies and their ethical
implications.

Aligning Zimbabwe's governance frameworks with international standards and best practices
in AI governance can facilitate knowledge sharing and improve the effectiveness of local
regulations. Creating dedicated regulatory bodies for AI governance can enhance oversight
and accountability. These bodies should be responsible for monitoring AI implementations,
ensuring compliance with ethical standards, and addressing grievances. Additionally,
fostering public awareness and dialogue around AI governance can help build trust among
citizens and encourage responsible AI use.

The ethical challenges associated with AI in Zimbabwe are significant and require robust
governance frameworks to mitigate risks. While existing policies and institutional
frameworks demonstrate a commitment to data protection and technology adoption, they lack
specificity and comprehensive measures to address the unique challenges posed by AI in
recruitment and selection. By developing targeted policies, engaging stakeholders, and
aligning with international standards, Zimbabwe can enhance its governance frameworks and
foster ethical AI practices that benefit society. The path forward involves not only addressing
the current gaps in governance but also embracing a collaborative approach that involves all
stakeholders in shaping the future of AI in Zimbabwe.

2.4 Empirical studies

The integration of Artificial Intelligence (AI) in Human Resources Management (HRM) has
transformed traditional practices, enhancing efficiency and decision-making (Leong, 2018).
However, it also raises significant ethical implications that merit thorough exploration. This
review synthesises empirical research on the ethical challenges and considerations associated
with AI in recruitment and selection.

The exploration of the ethical implications of using artificial intelligence (AI) in human
resources management (HRM) has garnered significant scholarly attention, revealing a
complex landscape of challenges and considerations. One prominent theme in the literature is
the issue of bias and discrimination. Barocas and Selbst (2016), in their study, found that
algorithmic bias can arise from training data that reflects societal inequalities, resulting in AI
systems that favour certain demographic groups over others. The algorithmic concern is
further echoed by Dastin (2018), who found that AI tools could disproportionately
disadvantage candidates from under-represented groups, raising ethical questions about
fairness and equity in hiring practices.

Bogen and Rieke (2018) examined algorithmic hiring tools and found that AI systems can
reflect past discriminatory practices since they rely on historical data. Bogen and Rieke
(2018) found that without robust oversight, AI systems tend to replicate societal inequalities,
hence undermining fairness in candidate selection. In the same vein, Cowgill (2018)
corroborated this by providing evidence to the effect that automated resume screening could
inadvertently filter out qualified candidates from certain under-represented groups. From the
outcomes gleaned from these two studies by Bogen and Rieke (2018) and Cowgill (2018), it
is apparent that they collectively stress the value of validating and standardising AI tools to
ensure they do not reinforce existing biases.

Another ethical dimension of study has been the area of transparency and accountability of
AI-driven recruitment and selection systems. Raghavan et al. (2020) focused on how the
“black box” nature of most AI-driven decision-making processes makes it difficult to hold
these systems accountable. The research study brings to the fore the fact that without
transparency, it becomes very difficult for key players or stakeholders to understand or even
challenge recruitment and selection decisions. Insights gleaned from the study by Raghavan
et al. are further corroborated by Zawisza and White (2020), who focused on a systematic
review of AI technologies. Their research study's key outcome was the conclusion that the
rapid development of AI technologies has outpaced the development of ethical guidelines.
Their recommendation was that there is a need for coming up with interdisciplinary ethical
frameworks that act as guides for organisations in maintaining accountability throughout all
stages of deploying AI.

Another critical area of concern is transparency in AI decision-making processes. Many AI

algorithms operate as "black boxes," making it difficult for HR professionals and candidates
to understand how decisions are made. Burrell (2016), in his study, emphasises that the lack
of transparency can lead to a mistrust of AI systems, as stakeholders may be unable to
ascertain whether hiring decisions are being made fairly or justifiably. This opacity
complicates the ethical landscape, as organisations may struggle to provide explanations for
AI-driven decisions, which is particularly concerning in situations involving hiring and
promotions.

Privacy and data protection also emerge as significant ethical considerations in the literature.
Recruitment and selection AI systems often require access to vast amounts of personal data to
function effectively, raising concerns about the potential for data breaches and misuse. Martin
(2019), in his study, argues that organisations must navigate the delicate balance between
leveraging data for AI applications and ensuring the privacy rights of individuals. The
General Data Protection Regulation (GDPR) in Europe has set a precedent for data protection
standards, yet many organisations in other regions, including developing countries, may lack
similar frameworks, leaving candidates vulnerable to privacy infringements when AI is
deployed in recruitment and selection contexts.

Furthermore, the ethical implications of AI in HRM extend to issues of accountability and

responsibility. As organisations increasingly rely on AI systems for critical HR functions such
as recruitment and selection, the question of who is accountable for AI-driven decisions
becomes paramount. If an AI system makes a biased hiring decision, should the blame fall on
the technology, the developers, or the HR professionals using it? Cath et al. (2017), in their
study, argue that establishing clear lines of accountability is essential for ethical AI
deployment, suggesting that organisations must develop governance frameworks that
delineate responsibilities related to AI usage.

The literature also emphasises the importance of ethical guidelines and frameworks to govern
the use of AI in HRM. Several scholars advocate for the development of ethical AI principles
that prioritise fairness, accountability, and transparency. For instance, Jobin et al. (2019)
propose a set of ethical guidelines for AI that organisations can adopt to ensure responsible
AI practices. These guidelines call for inclusive stakeholder engagement in AI development,
ongoing monitoring of AI systems for bias, and the establishment of clear governance
structures to oversee AI use in organisations.

Moreover, the role of organisational culture in shaping the ethical implications of AI in HRM
is increasingly recognised. Studies indicate that organisations with strong ethical cultures are
more likely to implement AI in ways that prioritise ethical considerations and employee well-
being. For example, a study by Kahn et al. (2020) found that organisations that foster open
communication and ethical behaviour among employees were better equipped to address the
challenges posed by AI technologies. This finding underscores the importance of cultivating a
culture that values ethics and encourages dialogue about the implications of AI in HRM.

The exploration of the ethical implications of using AI in recruitment and selection reveals a
multifaceted landscape characterised by concerns about algorithmic bias, transparency,
privacy, accountability, and the need for ethical guidelines. The literature underscores the
importance of addressing these ethical challenges proactively to ensure that AI technologies
are implemented in ways that promote fairness and accountability. By fostering a strong
ethical culture and developing comprehensive governance frameworks, organisations can
navigate the complexities of AI integration in recruitment and selection while mitigating the
associated risks and maximising the benefits of technological advancement.

2.5 Theoretical framework

The exploration of the ethical implications of using artificial intelligence (AI) in human
resource management (HRM) necessitates a robust theoretical framework that integrates
various ethical theories and principles. This framework is essential for understanding the
multifaceted ethical challenges posed by AI technologies in HRM as well as recruitment and
selection functions, including issues related to bias, transparency, accountability, and the
impact on employee rights and dignity. The following theoretical perspectives provide a
comprehensive foundation for analysing AI adoption and its ethical implications.

2.5.1 The information processing theory of organisations (Galbraith, 1973)

According to Malone (2019), researchers have continuously studied the impact of artificial
intelligence on organisational performance in the previous decade. It then illustrates an
increasing prominence of AI and HRM functions that include recruitment and selection, the
focus area for this research study. The common notion, according to Malone (2019), is that AI
allows a firm to adjust its employees to the technological volatiles in the world of work.
Galbraith's (1973) revolutionary 'information processing theory of organizations’ concept has
been the backbone for such. Daft (2019) gives details of the theory, saying it posits that
organisations function as systems for processing information, and the key to success is to
have efficient information processing capabilities. Within this framework, AI is viewed as a
tool that can enhance organizations' ability to process information with greater efficiency and
effectiveness, ultimately resulting in enhanced performance. This theory focuses on how
organisations internally acquire, store, retrieve, and distribute information to achieve their
goals. Galbraith’s (1973) organisational concept aims to disrupt traditional human resource
strategies by establishing artificial intelligence as an information processing capability. The
researcher suggests that automated firms address diverse HR requirements inclusive of
recruitment and selection. Because the implementation of artificial intelligence ensures that
an organisation remains relevant and competitive in the market, it makes Galbraith’s theory
primarily applicable to the study. According to Lawler and Berrin (2019), technological
advancement is pushing companies to adopt new tools, thereby transforming the workplace
and influencing the nature of work itself. In their article they highlight the importance of
technology in today's business world, pointing to the growing use of Artificial Intelligence
and Human Resource Information Systems as a result. The information processing theory of
organisations is applicable to the research because it recognises the dynamic information
capability processes that occur in the workplace, making AI an essential information
capability that can be used in the processes to achieve efficient organisational performance.

The Information Processing Theory offers insights into how recruiters process candidate
information, revealing cognitive biases that may be reinforced by AI systems (Binns, 2018).
It examines how AI tools analyse data for decision-making, promoting transparency and
fairness in recruitment (O'Neil, 2016). The theory also highlights the interaction between
recruitment professionals and AI, identifying ethical dilemmas from over-reliance on
automation (Crawford, 2016). Additionally, it informs the development of feedback
mechanisms in AI systems, enabling them to learn and improve fairness and accuracy in
hiring practices (Holmes, 2018). Overall, the theory aids in exploring the ethical implications
of AI in recruitment, fostering more equitable hiring processes.

However, the focus of this study is on the ethical implications of AI adoption in recruitment
and selection, and the researcher acknowledges that the applicability of the information
processing theory of organisations may be limited in terms of its practical traits, like its lack
of sociomaterial practices (Orlikowski, 2019). He argues for a sociomaterial perspective. This
approach recognises that AI is not just a tool but a technology that interacts with the social
fabric of an organisation. Information processing theory’s focus on information processing
alone neglects these social and cultural aspects of AI integration; hence the need to include
other theories of ethics.

2.5.2 Utilitarianism (Jeremy Bentham, 18th century)

Utilitarianism, a consequentialist theory, posits that the ethical value of an action is

determined by its outcomes, specifically the extent to which it promotes overall happiness or
well-being (Mill, 1863). In the context of AI in recruitment and selection, utilitarianism can
be applied to evaluate the benefits and drawbacks of AI technologies. For instance, AI can
enhance recruitment processes by improving efficiency and reducing biases in candidate
selection, potentially leading to a more diverse workforce (Binns, 2018). However, if AI
systems inadvertently perpetuate existing biases or lead to job displacement, the overall
utility may be compromised. Therefore, a utilitarian approach encourages organisations to
weigh the positive outcomes of AI implementation against potential negative consequences,
ensuring that the net benefits to the organisation are maximised. For the study, this theory
denotes that if AI use does greater good, then its use is justified, and the reverse is true.
 2.5.3 Conceptual Framework

Figure 2.5.3 Conceptual framework

Artificial Intelligence
in Recruitment and
Selection

Transparency and
Algorithmic Bias and Fairness Data Privacy and Consent Accountability

Ethical Implications
 Discrimination Risk
 Breach of Privacy
 Trust Deficiency

Governance Frameworks
(Internal and External)

Regional & National Laws &

Global Standards (EU AI Act, Policies (Zim Cyber and Data Organisational Practices (ISO
GDPR, OECD) Protection Act, Labour Act 27001, ISO AI)
(Cap) 28:01, AU Provisions)

Outcomes
 Ethical AI Use
 Fair Hiring Practices
 Enhanced Trust
Source: Researcher (2025)
The conceptual framework for this study shown above explores the complex relationship
between Artificial Intelligence (AI), ethical challenges, and governance in the context of
recruitment and selection. At the centre of the framework is the growing use of AI tools in
recruitment processes, such as automated resume screening, AI-driven interviews, and
predictive analytics. While these technologies enhance efficiency and reduce human bias,
they also introduce a set of significant ethical challenges that require thoughtful exploration.

Among the foremost ethical concerns is algorithmic bias, where AI systems may
unintentionally replicate or reinforce existing prejudices embedded in historical hiring data.
This can result in discriminatory hiring practices that favour certain demographic groups over
others, undermining principles of fairness and equal opportunity. In addition, the use of AI in
recruitment typically involves the collection and analysis of vast amounts of personal data.
This raises important issues around data privacy and informed consent, as candidates may not
be fully aware of how their information is being used, stored, or shared. A further concern is
the lack of transparency and accountability in many AI systems. Often described as “black
boxes,” these systems do not easily allow HR practitioners or candidates to understand how
decisions are made or to challenge those decisions when necessary.

These ethical challenges give rise to a range of implications. When AI is not properly
managed, it can lead to unjust discrimination, breaches of data privacy, and diminished trust
in recruitment processes. Over time, this may damage an organisation’s reputation and its
ability to attract diverse talent. It also raises critical questions about the legitimacy of
decisions made by systems that cannot be explained or held accountable. As a result, the
ethical use of AI in recruitment and selection cannot be left to chance. It must be carefully
regulated and guided by appropriate governance frameworks.

Governance mechanisms play a central moderating role in this framework. These operate at
multiple levels. Globally, frameworks such as the General Data Protection Regulation
(GDPR), the European Union AI Act, and ISO/IEC standards set out clear expectations for
transparency, fairness, and accountability in AI use. At the national level, the Zimbabwean
Cyber and Data Protection Act provides guidelines for informed consent, data protection, and
the rights of data subjects, even though it does not yet speak directly to AI systems. The
Labour Act (Chapter 28:01), although not AI-specific, prohibits unfair discrimination in
hiring. Organisational governance, such as the adoption of the ISO 27001 standard at Cimas
Health Group, helps ensure that data security and risk management practices are in place.
When applied effectively, these frameworks can mitigate the ethical risks of AI and support
its responsible integration into recruitment and selection processes.

Ultimately, the framework suggests that the ethical implications of AI in recruitment and
selection can be managed through a combination of strong internal policies and alignment
with national and international regulations. Where governance is effective, organisations are
more likely to experience positive outcomes such as fair and transparent hiring, ethical use of
technology, increased trust among stakeholders, and enhanced compliance with legal and
ethical standards. This conceptual framework thus offers a robust foundation for examining
how AI can be harnessed in recruitment in ways that are not only efficient but also ethically
sound and socially responsible.

2.5.4 Research gap

The integration of Artificial Intelligence (AI) into recruitment and selection has gained
significant attention in recent years, yet there remains a substantial research gap regarding the
ethical implications associated with this technological advancement. While existing literature
has explored various facets of AI in recruitment, there is a distinct lack of comprehensive
studies that focus specifically on the ethical challenges that emerge from these applications.
One of the primary gaps is the insufficient exploration of bias and discrimination in AI
algorithms used for recruitment processes. Research has shown that AI systems can
perpetuate existing biases present in training data, leading to unfair treatment of certain
demographic groups (Dastin, 2018; Binns, 2018). Additionally, the ethical implications of
data privacy and consent in AI-driven recruitment and selection practices have not been
sufficiently addressed. As organisations increasingly utilise personal data to train AI systems,
concerns about data security and individual privacy rights become paramount (Smith, 2021).
Current literature tends to focus on the technical aspects of data handling rather than the
ethical considerations surrounding informed consent and the potential risks of data misuse,
thus leaving a significant gap in understanding how organisations can ethically navigate these
challenges. Furthermore, while some studies have highlighted the importance of transparency
in AI decision-making, there is a lack of empirical research investigating how transparency
can be effectively implemented in recruitment and selection practices. The concept of the
"black box" in AI systems raises questions about accountability and the ability of HR
professionals to explain AI-driven decisions to candidates (Chui et al., 2019). However,
empirical evidence on best practices for enhancing transparency and fostering trust between
employees and AI systems in recruitment and selection is limited. Additionally, there is a
scarcity of research examining the regulatory landscape surrounding AI in recruitment and
selection within specific geographical contexts, particularly in developing countries like
Zimbabwe. The ethical implications of AI deployment may vary significantly based on local
laws, cultural norms, and socio-economic conditions (Munyoro, 2021). However, much of
the current discourse on AI ethics is dominated by Western perspectives, which may not
adequately address the unique challenges faced by organisations in different regions. This
lack of contextualised research limits the applicability of existing ethical frameworks and
guidelines for AI in recruitment and selection.

The ethical implications of using AI in recruitment and selection present a multifaceted

research gap that encompasses bias, data privacy, transparency, contextual differences, and
candidates’ well-being. Addressing these gaps is essential for developing a comprehensive
understanding of how organisations can ethically navigate the integration of AI technologies
into their recruitment and selection practices. Future research should aim to fill these gaps by
employing empirical methods, engaging diverse perspectives, and considering the unique
challenges faced by organisations in different contexts. By doing so, scholars and
practitioners can contribute to the creation of ethical frameworks and best practices that
ensure the responsible use of AI in recruitment and selection, ultimately benefiting both
organisations and candidates alike.

2.6 Chapter summary

The chapter mainly discussed the literature that surrounds the matters of ethical implications
in artificial intelligence adoption as an initiative to reshape recruitment and selection
strategies, including an analysis of ethical theories used to encourage ethical adherence in the
adoption and implementation of artificial intelligence. As the chapter unfolded, it exposed
how AI is connected to the overall performance of an organisation as well as its HR practices
around the area of recruitment and selection and how the influence of artificial intelligence
can benefit the organisation together with the ethical implications. The following chapter,
chapter 3, will focus on presenting the research methodology adopted for this study.