Next Article in Journal
Bridging Sustainability and Performance: Conceptualizing Net-Zero Integration in Construction Supply Chain Evaluations
Previous Article in Journal
Spatiotemporal Evolution and Driving Factors of NPP in the LanXi Urban Agglomeration from 2000 to 2023
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 17
Issue 13
10.3390/su17135815
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Perceptions of the Impact of AI on Human Resource Management Practices Among Human Resource Managers Working in the Chemical Industry in Saudi Arabia

by
Saeed Turki Alshahrani
*,
Jamel Choukir
,
Saja Albelali
and
Abdulaziz Abdulmohsen AlShalhoob
College of Business, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh 11432, Saudi Arabia
*
Author to whom correspondence should be addressed.
Sustainability 2025, 17(13), 5815; https://doi.org/10.3390/su17135815
Submission received: 18 May 2025 / Revised: 22 June 2025 / Accepted: 23 June 2025 / Published: 24 June 2025
(This article belongs to the Section Sustainable Management)
Browse Figures
Versions Notes

Abstract

The objective of this study is to investigate perceptions among HR managers in Saudi Arabia and compare these perceptions across demographic characteristics. Furthermore, the study examines the influence of AI knowledge and frequency of use on perceptions. An online survey was administered to a purposive sample of 420 HR managers working in the chemical industry in Saudi Arabia, and 234 complete responses were received. Data were analyzed using descriptive statistics, one-way ANOVA, and structural equation modeling. Findings show that AI was perceived positively, particularly in salary management, recruitment, performance evaluation, and training, but there were concerns about the loss of jobs and privacy. HR managers with higher education had a higher positive perception towards recruitment, selection, training, and performance appraisals. Knowledge and frequency of AI use had a positive influence on performance appraisal, recruitment and selection, and training, but had no influence on compensation and rewards. This study contributes to the literature by investigating perceptions of HR managers in the Saudi Arabia context. This is especially relevant in the context of technological advancement and Vision 2030 ambitions. Specifically, AI has the potential to create a skilled workforce eager for green innovation.

1. Introduction

Artificial intelligence (AI), rooted in disciplines such as philosophy, mathematics, computation, psychology, neuroscience, biology, and linguistics, has become indispensable in the services and manufacturing sectors [1,2]. AI can be interpreted from various perspectives. Some authors perceive it as an opportunity or a threat [3,4,5], while others see it as a threat and an opportunity [6]. Specific opportunities in human resource management include predicting workplace hazards, assessing and predicting employee satisfaction, optimizing work scheduling, and forecasting employee turnover [7,8].
Threats include job losses, a lack of trust in intelligent technologies among employees and managers, and reduced employee commitment as well as productivity [9,10]. Conversely, ref. [11] argues that AI does not present competition with machines but rather fosters a symbiotic relationship between humans and machines.
Ref. [6] supports the notion of a symbiotic relationship, arguing that viewing AI as a panacea is shortsighted. Instead, potential partnerships between AI and human resource management should be investigated. Murugesan et al. [12] observe that, although AI is replacing many tasks traditionally performed by HRM, flexibility remains essential for effective talent management. AI plays a critical role in providing this flexibility and enhancing agility in HRM.
Ref. [13] defines AI as a field of science and engineering dedicated to developing intelligent systems, comprising software and hardware designed to accomplish specific objectives. Ref. [14] describes AI as a scientific discipline centered on intelligent behavior, aiming to replicate human capabilities such as thinking, sensing, and reacting. Ref. [15] characterizes AI as a system’s ability to accurately interpret external data, learn from it, and apply those insights to achieve specific goals. Ref. [16] defines AI as an engineering technology that processes information by emulating human cognitive activities. AI systems are designed to think and act like humans or to exhibit rational thought and behavior [15,17]. These AI technology applications enable managers and employees to streamline or automate complex manual tasks. Ultimately, AI aims to create machines capable of performing tasks traditionally executed by humans. Big data and smart sensors are critical technologies for AI and its successful implementation in human resource management [18,19,20].
In a study on artificial intelligence, ref. [21] reported that 44% of manufacturing respondents viewed AI as crucial to the “production function”, while 49% deemed it “critical to success”. PricewaterhouseCoopers predicts the potential contribution of AI to reach USD 15.7 trillion by 2030 [22]. In Saudi Arabia, the government expects AI to generate USD 200 billion in revenue by 2035 [23].
The impact of AI on human resource management has been explored in multiple studies. Ref. [24] found that implementing AI in the HRM of a healthcare facility in the UAE required substantial capital investment but yielded competitive advantages. Specifically, AI accelerated the hiring process and facilitated the recruitment of better-qualified staff. Ref. [25] found that AI adoption in HRM across UAE companies reduced administrative tasks and minimized human bias. However, AI faced challenges such as lack of trust and the absence of a human touch, which are limitations not present in conventional hiring methods. Ref. [12] found that AI is vital for enhancing employee health and safety, as well as monitoring emotions and truancy, among IT, manufacturing, and services firms in Chennai and Bengaluru. Despite these significant benefits of AI in HRM, some regions continue to lag in its adoption. Ref. [26] notes that countries in the Global South, encompassing Africa, Asia, Latin America, and the Middle East, lag in AI adoption compared to those in the Global North. Human resource practitioners in Middle Eastern countries such as Jordan, Kuwait, Saudi Arabia, and Qatar exhibit a positive attitude toward implementing AI in human resource management [27,28]. Although these Middle Eastern countries have begun adopting AI in HRM, few studies have explored perceptions of HR managers on the impact of AI. Specifically, no study has examined perceptions of HR managers in Saudi Arabia. This research gap inspired the present study. The specific research questions investigated are as follows:
  • What are the perceptions of HR managers on the impact of AI on human resource management practices?
  • Are perceptions of HR managers different across age, education, and total working time?
  • How do knowledge and frequency of AI use influence perceptions of the impact of AI on human resource management practices?
Given the identified research gaps, this study contributes to existing knowledge in three key ways. First, it explores the perceptions of HR managers. Second, it addresses a research gap specific to the context of Saudi Arabia. Third, it enriches the literature by investigating the influence of AI knowledge and frequency of use on perceptions.
This paper is structured as follows: Section 2 reviews the literature on the main constructs and their interactions to establish a suitable theoretical framework. Section 3 outlines the research methodology. Section 4 and Section 5 presents the results and discussion. Finally, Section 6 summarizes the study, discusses its implications and limitations, and proposes directions for future research.

2. Literature Review

2.1. Theoretical Framework

The diffusion of innovation (DOI) theory argues that the adoption of new technologies is higher among individuals with a higher level of innovativeness [29]. This theory identifies four categories of adopters, as shown in Figure 1. The first group are innovators who adopt new technologies as they are “unconstrained by social norms”. The second group are early adopters who adopt technology because it offers more advantages than disadvantages. The third group are the late majority who are pressured to adopt technology. The fourth group are laggards who require formal evidence to adopt technology [30]. In the context of AI in HRM, countries in the Global North are early adopters and the late majority. Countries in the Global South such as Saudi Arabia are the late majority and laggards.
There are five critical variables in the diffusion theory that are relevant to the implementation of AI in HRM. Perceived compatibility is the extent to which users trust AI technologies in an environment where technology is not consistent with the current practices and needs of users. Technology that has higher compatibility and can increase the productivity of users will have higher adoption [31]. Triability is the extent to which users trust innovations and are willing to use AI applications [32]. Complexity is the level of difficulty users face, while relative advantage is the level of conviction that AI technology is better than existing practices [32].
There are several studies that have used DOI to investigate the adoption of AI in HRM. Ref. [33] found that the adoption of AI was determined by transformational leadership, intellectual stimulation, and leadership vision. Ref. [34] found that “compatibility, triability, and observability” had a positive relationship with the diffusion of AI in HRM but that complexity had no effect among HR professionals working in the pharmaceutical industry in Jordan. Ref. [35] found that the perceived ease of use and HR readiness had the highest impact on AI adoption among HR professionals in India. Ref. [36] found that, in the Tanzanian context, “relative advantage, compatibility, and competitive pressure” were the critical factors that determined AI adoption. Ref. [37] carried out a systematic review that revealed that the key themes that influenced AI adoption were operational, organizational, environmental, opportunities, challenges, and economic factors.
In the Saudi Arabian context there are a few studies that have investigated AI adoption in HRM. Ref. [38] found that the technology skills of employees were critical in the success of AI in HRM. Ref. [39] found that employee resistance, data security concerns, and costs were major challenges, but that there were opportunities from visionary leadership and company culture. Ref. [40] found that positive beliefs among HR managers led to a higher likelihood of AI adoption, while a negative attitude led to lower adoption willingness.
With DOI as the theoretical framework, we argue that greater knowledge of and skills in AI will have a positive influence on perceptions of the use of AI in recruitment and selection, training, compensation and rewards, career development, and performance appraisal. The suggested research model is illustrated on Figure 2.

2.1.1. Applications of AI in HRM

AI’s role in HRM is being studied across a diverse set of countries, such as Bahrain, China, Turkey, Malaysia, Morocco, and Bangladesh (Table 1). The mix of conceptual work and empirical studies shows a field that is still establishing itself, with some promising data starting to cement it [5,41]. The tilt toward qualitative over quantitative research makes sense too. The recurring focus on recruitment and selection, highlighted by [42] and others, feels like a natural starting point for AI. It is where data-crunching and pattern-spotting can shine, such as in sifting through heaps of applications or predicting candidate fit. There are also uses in training and development, performance appraisal, and pay in the future, which tracks with what ref. [4] states about AI touching every corner of HR. Some studies match specific AI tools to these tasks [4,20,43].

2.1.2. Benefits of AI in HRM

AI offers economic, financial, managerial, human, and strategic benefits when used in HRM.
Economic and Financial Benefits
AI aims to make users’ lives more efficient, improve living standards, and open new horizons for humanity [5,51]. AI can solve complex problems in a short time and provide lasting solutions. When using AI technology, data and information are well protected. Information or data stored in AI systems can be easily disseminated to large numbers of users. Additionally, AI technology can be relatively inexpensive. However, compared to human intelligence, training individuals and leveraging their natural intelligence is a costly, slow process [52]. Unlike humans, AI technologies generally do not exhibit unstable behavior and tend to have consistent performance. The financial consequences of AI adoption often involve more widespread decision-making based on cost–benefit principles.
Managerial Benefits
AI technology can enhance coordination and foster strong connections among employees handling different tasks [53]. Ref. [54] notes that decisions are often more accurate when AI is integrated into the decision-making process. Algorithmic decisions can be presented more quickly, objectively, and accurately compared to human decision-making, which is often based on experience and intuition. AI systems first collect information about subjects, events, and processes, analyze this data, and then execute their actions accordingly [55].
Strategic Benefits
An interesting evolution of HR departments, characterized by the shift from day-to-day operations to strategic decision-making, empowers HR to play a more pivotal role in shaping an organization’s future. By focusing on long-term outcomes rather than just being reactive, HR can help companies anticipate challenges and seize opportunities, which can be considered a significant transformation.
Human Benefits
AI is reshaping HR in a tangible way. The idea that it can replace outdated systems with faster and more-efficient ones highlights its potential to revolutionize the field. Ref. [43] predicts a 71% drop in recruitment costs and tripled productivity. This will happen in three ways: First, the ability to spot patterns in data that humans might miss is a significant opportunity, especially in a cost-conscious environment where every efficiency counts. Second, AI will free up time for HR staff to connect more with colleagues. Third, AI will take on routine tasks while leaving humans to focus on areas that need creativity, emotional depth, and savvy social skills. The perspective from [56], “let AI handle the predictable, codifiable tasks so people can tackle the messy, nuanced challenges that require judgment and imagination”, summarizes it well. AI becomes a support tool, clearing the deck for employees to focus on what truly matters. Criticisms of AI’s lack of human qualities, such as emotions or social skills, do not become a flaw; rather, they complement what people bring to the table. Machines can crunch data and screen resumes, but it is the human touch that seals the deal with top candidates. AI is moving into more complex, cognitively demanding tasks. The deployment of AI in HRM reduces the likelihood of subjective biases such as nepotism and favoritism in the recruitment and selection process. Additionally, AI has the potential to positively impact employee development, retention, and productive utilization.
It is clear that AI’s role in HR is expanding fast, touching everything from recruitment to workload management [46,57]. It is impressive how it not solely concerns efficiency anymore, but also unlocks new possibilities for how companies operate. The way that AI can streamline data collection and analysis, for instance, gives HR teams capabilities for turning raw data into actionable insights efficiently.
The call from [49] for more research is spot on. There is so much potential still untapped, and figuring out how AI reshapes an HR manager’s job or even HR’s broader role in a company could be a game-changer.

2.1.3. Challenges and Limitations of AI in HRM

The use of AI in HRM poses strategic, ethical, and legal challenges.
Strategic and Organizational
The perspective from [58] captures the excitement and uncertainty around AI in HR. The question of whether AI will boost productivity or disrupt jobs is significant, as noted by [3], and it is important to frame it as a challenge and an opportunity; it is not black and white, which makes it more intriguing. The argument from [59] about HR professionals’ readiness, hinging on cognitive, affective, and behavioral attitudes, is important. It is not just about the technology, as how people think and feel about AI also matter. If HR managers are anxious or skeptical, that could slow things down, no matter how powerful the tools are. The human element of this equation is just as critical as the algorithms. Ref. [6]’s argument of humans staying essential in complex, uncertain situations is valid.
Ethical and Legal
The idea of “ethical task-technology fit” makes sense, as the risks depend on what AI is required to accomplish and how well it aligns with these requirements [43]. When AI is given more decision-making roles, such as picking candidates or evaluating performance, it is not just a tech upgrade: it is a whole new ballgame. The risk of discrimination creeps in, whether intentional or not, and is a real concern, as AI can inherit biases from the data it is trained on [54]. The legal implications are also significant. Existing laws might not be sufficient when machines are making more decisions, and figuring out who is accountable, an algorithm, a programmer, or a company, becomes difficult quickly. New frameworks, policies, and contracts are necessary to keep things fair and transparent, especially around sensitive areas like workers’ privacy [32].
AI limitations in HR are like a three-legged stool, with human, ethical, and legal challenges holding it up. While AI can sift through resumes at lightning speed, it is the human touch that convinces candidates to sign. Ethical risks like data privacy and misuse have been noted by ref. [5], while ref. [60] points out risks of bias and discrimination. That “cognitive biases in biased AI” can exist is chilling, as it shows how human flaws can sneak into AI, amplifying issues like unfair hiring. Furthermore, there are challenges related to high costs, a lack of know-how, and the lingering fear of job losses [61]. It is easy to see why some might view AI as a threat when it is still so new and misunderstood. Ref. [57] argues AI could accidentally elbow humans out of the driver’s seat if not managed correctly. Furthermore, refs. [62,63] point out the creativity gap that can arise when AI is good at optimizing what is already there, but it is not dreaming up the next big idea or navigating the messy, emotional side of work. Ref. [64]’s emphasis on human intelligence stepping in where AI stumbles is essential, as humans and AI are complementary.
Technical and Operational
AI in HR is not just a tech challenge; it is a global puzzle with ethical and cultural pieces that do not always fit together neatly. The argument from ref. [65] about ethical standards varying across countries really highlights this. What is acceptable in one place might be a total no-go somewhere else. It is like trying to write a universal rulebook when everyone is playing by different rules. Cultural collaboration is one approach to bridge that gap by finding a common ground without forcing every nation to agree on every detail. The argument by [66], that AI needs to be seen as an opportunity instead of a boogeyman, is valid. It is a mindset shift that could unlock a lot of potential if people lean into it and focus on how AI can solve problems rather than just worrying about what it might break. Tying it back to academia’s role, as ref. [65] does, is a great call. Universities and researchers could be the glue, fostering that cross-cultural dialog and determining where compromise works and where it does not.
The idea of tailoring regulations to fit an organization’s ethical standards, rather than just borrowing from somewhere else, is a practical way to make AI work without clashing with local values. The emphasis of ref. [65] on respecting differences instead of steamrolling them with imposed rules is sensible. Collaboration does not imply domination. Authors such as those of refs. [66,67] highlight the ethical minefield, privacy, fairness, accountability, and human rights aspects of AI. It is concerning how AI could amplify bias or erode fairness if it is not handled with care, especially in HR where decisions affect people’s livelihoods. Other concerns raised by ref. [68], such as bias, discrimination, unemployment, data security, and accountability, read like a checklist of everything that could go wrong (Figure 3).
With the growing use of technology, AI has become essential in the digital age. However, AI presents issues and challenges, particularly in security and information protection. For example, attackers can exploit AI systems to create phishing emails for malicious purposes [69].

3. Research Methodology

3.1. Data Collection

Data was collected through an online questionnaire. From a pilot study of 20 human resources managers, a Cronbach’s alpha of 0.85 was observed. This indicates that the questionnaire had a high level of internal consistency. Insights from the pilot study were used to refine the questionnaire. After verification, a Google Forms survey was created and purposively sent to a sample of 420 Saudi HR managers working in chemical firms in the private sector. This purposive sampling ensured the selection of relevant participants from the target population. Of the 420 managers, 234 completed the survey, which was a response rate of 55.7%.

3.2. Instrument

A questionnaire consisting of three sections was used. The first section captured the demographic characteristics of participants, while the second section captured knowledge and the frequency of use of AI. The third section consisted of five dimensions adapted from [5] that captured perceptions of AI in human resource management practices. The compensation and rewards (PAICR) dimension consisted of four items. The career development (PAICD) dimension consisted of six items. The performance appraisal (PAIPA) dimension consisted of six items. The recruitment and selection (PAIRS) dimension consisted of nine items. The training (PAIT) dimension consisted of five items. Responses were collected using a five-point Likert scale (strongly agree, agree, neutral, disagree, and strongly disagree) selected for its user-friendliness.

3.3. Data Analysis

Frequencies and percentages were used to summarize demographic characteristics as well as the knowledge and the frequency of AI use. Mean, median, skewness, and kurtosis were used to summarize items in each dimension. A composite score consisting of the mean of items in each dimension was calculated. A one-way ANOVA was used to examine differences in the composite score of each dimension across age, education, and total working time. Post hoc comparisons were carried out using the least significant difference (LSD). Structural equation modeling (SEM) was used to investigate the effect of the knowledge and frequency of use dimension on each of the dimensions of human resource management practices. SEM was selected for two reasons: First, SEM was preferred over regression as it provides more detail on the relationship between all the variables in a model [70]. Second, there are consistent algorithms and better criteria that can be used to assess discriminant validity [71]. Model fit was assessed using CFI, TLI, SRMR, and RMSEA. Construct reliability was assessed using Cronbach’s alpha and composite reliability. AVE was used to assess convergent validity. HTMT and the Fornell and Larcker criterion were used to assess discriminant validity. Data were analyzed using R-4.4.3 software. Specifically, the lavaan package was used for SEM.

4. Results

4.1. Demographic Characteristics

Demographic characteristics of participants are summarized in Table 2. The sample is predominantly male (79.5%), which may skew perceptions if gender influences opinions on AI in HRM. The low female representation (20.5%) suggests the potential underrepresentation of female perspectives. The 30–34 age group is the largest (32.1%), followed by the 40–44 age group (25.6%). Younger respondents (25–39) dominate (71.7%), suggesting that the sample reflects the perspectives of early- to mid-career professionals, who may be more open to technological changes like AI. Most respondents are university graduates (56.4%) or postgraduates (28.2%), totaling 84.6%. This highly educated sample may have a better understanding or acceptance of AI technologies. Respondents have varied experience, with 6–10 years of total working time being the most common (33.3%). Many are relatively new to their current organizations (0–5 years: 33.3%) and HRM roles (0–5 years: 42.3%), indicating a mix of fresh and seasoned perspectives on AI’s role in HRM.
Table 3 shows a summary of AI knowledge and frequency of use among respondents. Most respondents have average (46.2%) or little (29.5%) AI knowledge, with only 14.2% reporting much or very much. This suggests limited exposure, which can negatively influence optimism about AI in HRM. AI usage is low, with 51.3% reporting very little or little use. This limited engagement may reflect early-stage AI adoption. AI use in HRM is even rarer, with 41.0% reporting very little and only 12.8% using it much or very much. This indicates AI is not yet widely integrated into HRM practices among respondents.

4.2. Perceptions of HR Managers on the Impact of AI on HRM

Descriptive statistics of items in each dimension are summarized in Table 4, Table 5, Table 6, Table 7 and Table 8. The criteria developed by [5] are used to interpret average scores. Compensation and rewards (M = 3.54), career development (M = 3.72), recruitment and selection (M = 3.70), performance appraisal (M = 3.41), and training (M = 3.63) had scores between 3.40 and 4.19. This suggests the perception of change towards AI is at a high level.
A correlation heatmap showing interactions between different survey responses is illustrated in Figure 4. Respondents who believe that AI will help determine fair salaries also strongly agree that AI can reduce human-related delays in payments and ensure the accurate calculation of bonuses and premiums. This suggests that AI is perceived as a tool for improving transparency and reducing biases in salary and compensation management. A strong correlation exists between AI helping employees identify necessary qualifications for their dream careers and assisting in career planning. Those who think that AI will help recognize employees deserving of promotions also believe it will automate wage rises based on skill increases. This indicates that respondents see AI as a career development enabler, helping them to acquire needed skills and opportunities for promotion. Those who think that AI will positively impact corporate culture through performance evaluation also believe that AI will accurately predict future employee performance. A strong link exists between AI ensuring motivation during performance assessments and AI determining fair performance evaluation criteria. This suggests that fair and transparent AI-based evaluation systems could enhance corporate culture and employee satisfaction. Respondents who believe that AI reduces time spent in finding candidates also think that it will increase access to qualified candidates and reduce stress in hiring decisions. Strong correlations exist between AI analyzing resumes in detail and AI selecting the most suitable personnel. These findings suggest that AI is widely perceived as a beneficial recruitment tool, streamlining hiring and improving selection quality. Those who believe AI will make training more accessible also agree that AI will reduce time spent on training and eliminate geographical constraints on learning. AI is perceived as improving learning retention and engagement, reducing attention deficits in traditional training methods.
AI will make training more accessible. Additionally, AI will reduce the time spent on training and eliminate geographical constraints on learning. AI is perceived as improving learning retention and engagement, reducing attention deficits in traditional training methods.

4.3. Differences in Perceptions of the Impact of AI Across Age, Education, and Total Working Time

From Table 9, it can be observed that the mean of the compensation and rewards (PAICR) dimension is not different across the age and total working time categories. However, the mean is different across the education categories. Post hoc comparison using LSD revealed that the means in the university graduate (M = 3.43), diploma (M = 3.42), and high school education (M = 3.21) categories were similar. However, the mean in the postgraduate (M = 3.89) category was significantly different. Cohen’s d indicated that differences between the postgraduate and university graduate categories were moderate (es = 0.61), those between the postgraduate and high school education categories were moderate (es = 0.66), and those between the postgraduate and diploma categories were moderate (es = 0.76). These results suggest that participants with postgraduate education had the highest perception of change towards compensation and rewards.
The mean of the career development (PAICD) dimension is not different across the age and education categories. However, the mean is different across the total working time categories. Post hoc comparison showed that all of the categories had significantly different means: 11–15 years (M = 3.92), 6–10 years (M = 3.80), 16–20 years (M = 3.73), more than 21 years (M = 3.55), and 0–5 years (M = 3.37). Cohen’s d indicated that differences between 0–5 years and 11–15 years were moderate (es = 0.7), differences between 0–5 years and 16–20 years were small (es = 0.45), differences between 0–5 years and 6–10 years were moderate (es = 0.56), differences between 11–15 years and 16–20 years were small (es = 0.26), and differences between 11–15 years and more than 21 years were small (es = 0.43). These results suggest managers that worked for the shortest period had the lowest perception of change towards career development.
The mean of the performance appraisal (PAIPA) dimension is not different across the age and education categories. However, the mean is different across the total working time categories. Post hoc comparison using LSD revealed that means in the 16–20 years (M = 3.46) and 6–10 years (M = 3.46) categories were similar. However, the means in the 11–15 years (M = 3.67), more than 21 years (M = 3.26), and 0–5 years (M = 3.00) categories were different. Cohen’s d indicated that differences between the 0–15 years and 11–15 years categories were large (es = 0.83), differences between the 0–5 years and 16–20 years categories were moderate (es = 0.61), differences between the 0–5 years and 6–10 years categories were moderate (es = 0.56), and differences between the 11–15 years and more than 21 years categories were moderate (es = 0.56). These results are similar to those for career development, as managers who had worked for the shortest time had the lowest perception of change. However, the means across the categories in the PAIPA dimension were lower compared to the PAICD dimension. This suggests that participants had a higher perception of change towards performance appraisal compared to career development.
The mean of the recruitment and selection (PAIRS) dimension is not different across the total working time but is different across the age and education categories. Post hoc comparison revealed that the age categories 25–29 (M = 3.87), 50 and above (M = 3.86), and 30–34 (M = 3.75) had similar means. However, the age groups 45–49 (M = 3.61) and 40–44 (M = 3.50) had significantly lower means. Cohen’s d indicated that differences between the 25–29 years and 40–44 years categories were moderate (es = 0.58), while the other pairwise differences were small or negligible. These results generally suggest that older managers have a lower perception of change towards AI. However, the cause of the high perception of change observed among managers aged 50 and above years is not clear. Post hoc comparison of education categories revealed that the mean across all the categories was significantly different for the postgraduate (M = 3.95), university graduate (M = 3.43), diploma (M = 3.42), and high school education (M = 3.21) categories. Cohen’s d indicated that differences between the diploma and postgraduate categories were large (es = 1.32), differences between the diploma and university graduate categories were large (es = 0.94), and differences between the high school education and postgraduate categories were moderate (es = 0.65), while the other pairwise differences were small. These results suggest that there is a trend, with higher education having a higher perception of change compared to a lower category. This is in contrast to compensation and rewards, where the high school education, diploma, and university graduate categories had a similar mean.
The mean of the training (PAIT) dimension is not different across age and total working time, but is different across the education categories. Post hoc comparison revealed that the means across the postgraduate (M = 3.82), university graduate (M = 3.62), high school education (M = 3.40), and diploma (M = 3.20) categories were significantly different. Cohen’s d indicated that differences between the diploma and postgraduate categories were large (es = 0.9), while differences between the diploma and university graduate categories were moderate (es = 0.66), and the other pairwise differences were small. This trend is similar to that observed in recruitment and selection, where participants with higher education had higher perception of change.

4.4. The Influence of Knowledge and Frequency of AI Use on Perceptions of the Impact of AI

Figure 5 illustrates SEM being used to investigate the influence of the knowledge and frequency of AI use. The oval shapes show the latent variables, while the rectangles show the observed indicators. The model is correctly specified, as there is no item connected to more than one latent variable. The single-headed arrows show factor loadings, while double-headed arrows show correlations among latent variables. All standardized factor loadings are higher than 0.5, indicating that each dimension explains more than 25% of the variation in each item [72]. AI knowledge and skills were weakly correlated with training, career development, and compensation and rewards, but was moderately correlated with performance appraisal. Correlations among HRM constructs were strong.
Construct reliability and convergent validity measures are shown in Table 10. The Cronbach’s alpha for each dimension was higher than the required threshold of 0.7 [73]. The composite reliability for each dimension was higher than the required threshold of 0.7 [74]. Therefore, construct reliability was achieved for each dimension.
The AVE values of the PAICR and PAIRS dimensions were marginally lower than the required threshold of 0.5, but the AVE values of the other dimensions were higher [75]. Despite the lower AVE values of the PAICR and PAIRS dimensions, the convergent validity of these dimensions was achieved, as composite reliability was higher than 0.6 [76].
Discriminant validity measures are shown in Table 11 and Table 12. Discriminant validity is assessed using the Fornell–Larcker criterion by comparing the square root of the AVE shown in bold in the diagonal with correlations of constructs in rows and columns [75]. Some Fornell–Larcker criterion values in the same rows and columns were higher than the diagonal value, hence discriminant validity was not entirely established. Specifically, the correlations between the PAICR and PAICD dimensions, as well as the PAIPA and PAICD dimensions, were of concern. However, all HTMT ratios were lower than the required threshold of 0.90 [77,78], as shown in Table 13. In assessing discriminant validity, HTMT is superior and more stringent compared to Fornell–Larcker [79,80]. Therefore, discriminant validity was achieved in all dimensions.
The model fit metrics shown in Table 11 indicate minor concerns with model fit. The RMSEA and SRMR are below the required threshold. However, the CFI and TLI values of 0.87 and 0.89 were marginally lower than the required threshold of 0.90 [81,82].
Path coefficients are shown in Table 14. The coefficient of the path between knowledge and use and the PAICD dimension is positive and statistically significant. This suggests that respondents with higher knowledge and frequency of using AI had a higher perception of change towards career development. Similarly, the coefficients of the paths along the PAIPA, PAIRS, and PAIT dimensions are positive and statistically significant. This suggests that participants with higher knowledge and frequency of using AI had a higher perception of change towards performance appraisal, recruitment and selection, and training. However, the path along the PAICR dimension was not statistically significant. This suggests that the knowledge and frequency of AI use had no influence on the perception of change towards compensation and rewards. Furthermore, the lower mean rating on the PAICR dimension suggests that there are concerns around compensation and rewards.

5. Discussion

5.1. Theoretical Interpretation and DOI Linkages

Most respondents agree that AI can streamline HR processes, such as wage calculations, candidate selection, and performance evaluation. Many participants believe that AI will help adapt salary systems, reduce human-related delays in wage payments, and ensure fairer salary calculations; however, there is some concern about AI automating wage increases solely based on skill increases. Respondents see AI as helpful in guiding career paths by suggesting required skills for promotions. There is strong agreement that AI can help employees acquire necessary skills and develop career plans. Many believe that AI can accurately measure performance and positively impact corporate culture. Most participants agree that AI can reduce the time and stress involved in finding candidates, improve candidate quality, and make hiring decisions more stable. There is optimism that AI can provide detailed resume analysis and better personnel selection. Respondents believe that AI will increase accessibility to training, reduce attention deficits, and save time. Many believe that AI-driven training will remove location restrictions, making training more flexible. Overall, respondents see AI as a positive force in HR, particularly in salary management, recruitment, performance evaluation, and training. In context of the DOI theory, this suggests that HR managers in Saudi Arabia have a positive perception towards the triability and compatibility of AI with training, recruitment and selection, and performance evaluation. This positive impact of compatibility and triability has been similarly observed among HR managers in Jordan and Tanzania [35,36]. Positive perception of and higher knowledge and skills in AI having a positive effect on perceptions towards career development, performance appraisal, and recruitment and selection are consistent with the finding that employees’ technical skills and attitudes are critical to the success of AI in HRM [38,40]. In the context of DOI, those employees with a positive attitude and AI competence could be considered early adopters. Such employees can become opinion leaders that make a strong case for the adoption of AI in HRM. A significant portion of participants are neutral, indicating either uncertainty or a lack of strong opinions about AI’s role in HR. A smaller fraction disagrees or strongly disagrees, suggesting skepticism or resistance to AI-driven HR processes. Concerns exist around overcompetitive performance measurement, which some see as outdated with AI. However, there is still some uncertainty and skepticism, mainly about automation replacing human judgment. In the context of DOI, these concerns suggest that there could be complexity, and some HR managers may not be convinced that AI is superior to traditional practices. This employee resistance and concerns in data security have been similarly observed by Ref. [39], but can be mitigated by corporate culture and visionary leadership. In the context of DOI, such employees can be considered the early and late majority. Opinion leadership from early adopters can be used to shape corporate vision and culture to help drive the adoption of AI in HRM in the Saudi context.

5.2. Practical and Policy Implications

Insights from this study have important implications for human resource management practices: First, organizations need to leverage strong support for AI in recruitment and training, but should address concerns in performance management and salary fairness, possibly through education, given the link between AI knowledge and positive perceptions. Second, hiring for top HR leadership positions needs to prioritize younger candidates who have the highest level of education. Third, top management needs to address concerns around compensation and rewards, as participants appeared skeptical. Fourth, AI training for HR professionals with certification requirements verified by the Ministry of Human Resources and Social Development (HRSD) should be established. Fifth, regular “human-in-the-loop” audits, requiring senior management verification of AI-generated HR recommendations, should be developed.

5.3. Study Limitations and Future Research

The findings of this study need to be interpreted in consideration of several limitations. First, just slightly more than half of targeted respondents completed the survey. Although this sample was adequate for SEM, there was a low number of female participants. This limited the comparison of perceptions between male and female participants. Future studies need to target more female participants to enable a meaningful gender comparison. Second, this study was limited to the Saudi Arabian context, and may not be generalizable to other countries. Third, the literature that compares perceptions across demographic characteristics and investigates the influence of the knowledge and frequency of AI use is extremely limited. This severely limited the comparison of the current findings to those in the previous literature. There is a need for more studies that investigate these aspects.

6. Conclusions

AI has emerged as a critical element in human resource management practices and has been adopted in countries in Global North. However, countries in the Global South are lagging in its adoption, despite the benefits that could be realized. This slow pace of adoption can be attributed to technological factors and perceptions among HR managers. Saudi Arabia is one of the countries lagging in the adoption of AI. Perceptions that could be hindering the adoption of AI in Saudi Arabia have not been investigated. It is this research gap that motivated this study. Data was collected from 234 HR managers. Results showed that HR managers in Saudi companies have a medium level of AI knowledge and a low level of AI usage. AI was perceived positively, particularly in salary management, recruitment, performance evaluation, and training. Specifically, AI is expected to increase access to training opportunities, reduce the time spent on training, and bridge geographical barriers in training. However, there is uncertainty in terms of AI replacing human judgment, overcompetitive performance management, and privacy. The comparison of perceptions across demographic characteristics revealed that HR managers with higher education had a higher perception of change toward recruitment, selection, training, and performance appraisals compared to career development. The knowledge and frequency of AI use had a positive influence on performance appraisal, recruitment and selection, and training, but had no influence on compensation and rewards.
The results of this study are relevant to the Saudi business environment, where Vision 2030 and rapid technological advancement are pushing organizations to modernize. HR leaders in this context may see AI as a means by which to align HRM with national goals of economic diversification and innovation, further reinforcing their positive outlook. However, limitations like cost, a lack of AI expertise, and resistance from employees accustomed to traditional practices could temper this enthusiasm, prompting a balanced view that embraces AI’s potential while recognizing the need for careful implementation.

Author Contributions

Conceptualization, J.C.; Methodology, S.T.A.; Investigation, J.C. and A.A.A.; Resources, S.A. and A.A.A.; Data curation, J.C.; Writing—original draft, S.T.A. and A.A.A.; Writing—review & editing, S.A.; Project administration, S.A. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University (IMSIU) (grant number IMSIU-DDRSP2504).

Institutional Review Board Statement

Our study complies with Imam Mohammad Ibn Saud Islamic University/Saudi Arabia guidelines, exempting our study from ethical review. The document from the Saudi Arabian National Committee of Bioethics, Article 13.5, indicates that it can be waived if the risk of research does not exceed minimal risk. The definition of minimal risk is also attached. https://ncbe.kacst.edu.sa/media/wpokc4ij/implementing-regulations-of-the-law-of-ethics-of-research-on-living-creatures_.pdf (accessed on 5 May 2025).

Informed Consent Statement

Informed consent for participation was obtained from all subjects involved in the study.

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Nabıyev, V.; Karal, H.; Arslan, S.; Erumıt, A.K.; Cebı, A. An Artificial Intelligence-Based Distance Education System: Artimat. Turk. Online J. Distance Educ. 2013, 14, 81–98. [Google Scholar]
  2. Ibarra, D.; Ganzarain, J.; Igartua, J.I. Business Model Innovation through Industry 4.0: A Review. Procedia Manuf. 2018, 22, 4–10. [Google Scholar] [CrossRef]
  3. Hogg, P. Artificial Intelligence: HR Friend or Foe? Strateg. HR Rev. 2019, 18, 47–51. [Google Scholar] [CrossRef]
  4. Bhardwaj, G.; Singh, S.V.; Kumar, V. An Empirical Study of Artificial Intelligence and Its Impact on Human Resource Functions. In Proceedings of the 2020 International Conference on Computation, Automation and Knowledge Management (ICCAKM), Dubai, United Arab Emirates, 9–10 January 2020; IEEE: New York, NY, USA, 2020; pp. 47–51. [Google Scholar]
  5. Kambur, E.; Akar, C. Human Resource Developments with the Touch of Artificial Intelligence: A Scale Development Study. Int. J. Manpow. 2022, 43, 168–205. [Google Scholar] [CrossRef]
  6. Jarrahi, M.H. Artificial Intelligence and the Future of Work: Human-AI Symbiosis in Organizational Decision Making. Bus. Horiz. 2018, 61, 577–586. [Google Scholar] [CrossRef]
  7. Davoudi Kakhki, F.; Freeman, S.A.; Mosher, G.A. Utilization of Machine Learning in Analyzing Post-Incident State of Occupational Injuries in Agro-Manufacturing Industries. In Advances in Safety Management and Human Performance, Proceedings of the AHFE 2020 Virtual Conferences on Safety Management and Human Factors, and Human Error, Reliability, Resilience, and Performance, San Diego, CA, USA, 16–20 July 2020; Springer: New York, NY, USA, 2020; pp. 3–9. [Google Scholar]
  8. Zhang, H.; Xu, L.; Cheng, X.; Chao, K.; Zhao, X. Analysis and Prediction of Employee Turnover Characteristics Based on Machine Learning. In Proceedings of the 2018 18th International Symposium on Communications and Information Technologies (ISCIT), Bangkok, Thailand, 26–29 September 2018; IEEE: New York, NY, USA, 2018; pp. 371–376. [Google Scholar]
  9. Raisch, S.; Krakowski, S. Artificial Intelligence and Management: The Automation–Augmentation Paradox. Acad. Manag. Rev. 2021, 46, 192–210. [Google Scholar] [CrossRef]
  10. Choi, D.Y.; Kang, J.H. Net Job Creation in an Increasingly Autonomous Economy: The Challenge of a Generation. J. Manag. Inq. 2019, 28, 300–305. [Google Scholar] [CrossRef]
  11. McAfee, A.; Brynjolfsson, E. Human Work in the Robotic Future: Policy for the Age of Automation. Foreign Aff. 2016, 95, 139–150. [Google Scholar]
  12. Murugesan, U.; Subramanian, P.; Srivastava, S.; Dwivedi, A. A Study of Artificial Intelligence Impacts on Human Resource Digitalization in Industry 4.0. Decis. Anal. J. 2023, 7, 100249. [Google Scholar] [CrossRef]
  13. Biliavska, V.; Castanho, R.A.; Vulevic, A. Analysis of the Impact of Artificial Intelligence in Enhancing the Human Resource Practices. J. Intell. Manag. Decis. 2022, 1, 128–136. [Google Scholar] [CrossRef]
  14. Gherghina, S.C. An Artificial Intelligence Approach towards Investigating Corporate Bankruptcy. Rev. Eur. Stud. 2015, 7, 5. [Google Scholar] [CrossRef]
  15. Haenlein, M.; Kaplan, A. A Brief History of Artificial Intelligence: On the Past, Present, and Future of Artificial Intelligence. Calif. Manag. Rev. 2019, 61, 5–14. [Google Scholar] [CrossRef]
  16. Popkova, E.G.; Sergi, B.S. Human Capital and AI in Industry 4.0. Convergence and Divergence in Social Entrepreneurship in Russia. J. Intellect. Cap. 2020, 21, 565–581. [Google Scholar] [CrossRef]
  17. Müller, V.C.; Bostrom, N. Fundamental Issues of Artificial Intelligence; Springer: New York, NY, USA, 2016; Volume 376, ISBN 3-319-26485-0. [Google Scholar]
  18. D’Arco, M.; Presti, L.L.; Marino, V.; Resciniti, R. Embracing AI and Big Data in Customer Journey Mapping: From Literature Review to a Theoretical Framework. Innov. Mark. 2019, 15, 102. [Google Scholar] [CrossRef]
  19. Bailey, D.E.; Barley, S.R. Beyond Design and Use: How Scholars Should Study Intelligent Technologies. Inf. Organ. 2020, 30, 100286. [Google Scholar] [CrossRef]
  20. Tuffaha, M. The Impact of Artificial Intelligence Bias on Human Resource Management Functions: Systematic Literature Review and Future Research Directions. Eur. J. Bus. Innov. Res. 2023, 11, 35–58. [Google Scholar] [CrossRef]
  21. Plathottam, S.J.; Rzonca, A.; Lakhnori, R.; Iloeje, C.O. A Review of Artificial Intelligence Applications in Manufacturing Operations. J. Adv. Manuf. Process. 2023, 5, e10159. [Google Scholar] [CrossRef]
  22. PricewaterhouseCoopers. PwC’s Global Artificial Intelligence Study: Sizing the Prize. Available online: https://www.pwc.com/gx/en/issues/artificial-intelligence/publications/artificial-intelligence-study.html (accessed on 16 June 2025).
  23. GlobeNewswire. The Kingdom of Saudi Arabia Artificial Intelligence (Ai) Market Is Expected to Reach Revenue of USD 61,854.4 Mn by 2033, at 46.6% Cagr: Dimension Market Research. Available online: https://www.globenewswire.com/news-release/2024/12/09/2993958/0/en/The-Kingdom-Of-Saudi-Arabia-Artificial-Intelligence-Ai-Market-Is-Expected-To-Reach-Revenue-Of-USD-61-854-4-Mn-By-2033-At-46-6-Cagr-Dimension-Market-Research.html (accessed on 16 June 2025).
  24. Li, P.; Bastone, A.; Mohamad, T.A.; Schiavone, F. How Does Artificial Intelligence Impact Human Resources Performance. Evidence from a Healthcare Institution in the United Arab Emirates. J. Innov. Knowl. 2023, 8, 100340. [Google Scholar] [CrossRef]
  25. Singh, A.; Shaurya, A. Impact of Artificial Intelligence on HR Practices in the UAE. Humanit. Soc. Sci. Commun. 2021, 8, 312. [Google Scholar] [CrossRef]
  26. Adekoya, O.D.; Mordi, C.; Ajonbadi, H.A. HRM, Artificial Intelligence and the Future of Work: Insights from the Global South; Springer Nature: Cham, Switzerland, 2024; ISBN 978-3-031-62368-4. [Google Scholar]
  27. Hmoud, B.; Varallyai, L. Role of Artificial Intelligence in Human Resource Management in the Middle East Countries. KnE Soc. Sci. 2023, 8, 435–448. [Google Scholar] [CrossRef]
  28. Pramod, D. Robotic Process Automation for Industry: Adoption Status, Benefits, Challenges and Research Agenda. Benchmarking Int. J. 2022, 29, 1562–1586. [Google Scholar] [CrossRef]
  29. Uzumcu, O.; Acilmis, H. Do Innovative Teachers Use AI-Powered Tools More Interactively? A Study in the Context of Diffusion of Innovation Theory. Technol. Knowl. Learn. 2024, 29, 1109–1128. [Google Scholar] [CrossRef]
  30. Dearing, J.W.; Cox, J.G. Diffusion of Innovations Theory, Principles, and Practice. Health Aff. 2018, 37, 183–190. [Google Scholar] [CrossRef] [PubMed]
  31. Venkatesh, V.; Morris, M.G.; Davis, G.B.; Davis, F.D. User Acceptance of Information Technology: Toward a Unified View. MIS Q. 2003, 27, 425–478. [Google Scholar] [CrossRef]
  32. Almaiah, M.A.; Alfaisal, R.; Salloum, S.A.; Hajjej, F.; Shishakly, R.; Lutfi, A.; Alrawad, M.; Al Mulhem, A.; Alkhdour, T.; Al-Maroof, R.S. Measuring Institutions’ Adoption of Artificial Intelligence Applications in Online Learning Environments: Integrating the Innovation Diffusion Theory with Technology Adoption Rate. Electronics 2022, 11, 3291. [Google Scholar] [CrossRef]
  33. Patnaik, P.; Bakkar, M. Exploring Determinants Influencing Artificial Intelligence Adoption, Reference to Diffusion of Innovation Theory. Technol. Soc. 2024, 79, 102750. [Google Scholar] [CrossRef]
  34. Obeidat, B.Y. The Relationship between Innovation Diffusion and Human Resource Information System (HRIS). Bus. Manag. 2013, 5, 72–96. [Google Scholar]
  35. Khan, F.A.; Khan, N.A.; Aslam, A. Adoption of Artificial Intelligence in Human Resource Management: An Application of TOE-TAM Model. Res. Rev. Hum. Resour. Labour Manag. 2024, 5, 22–36. [Google Scholar]
  36. Faustine, P.; Rachmawati, R. AI Adoption Determinants and Its Impacts on HRM Effectiveness within MES in Tanzania. Open J. Bus. Manag. 2024, 12, 2532–2552. [Google Scholar] [CrossRef]
  37. Pedrami, M.; Vaezi, S.K. Factors Influencing Artificial Intelligence Adoption in Human Resource Management: A Meta-Synthesis and Systematic Review of Multidimensional Considerations. J. Work.-Appl. Manag. 2025. ahead of print. [Google Scholar] [CrossRef]
  38. Al-Ayed, S. Role of Artificial Intelligence for Strengthening Human Resource System via Mediation of Technology Competence. Probl. Perspect. Manag. 2024, 22, 518. [Google Scholar] [CrossRef]
  39. Madanchian, M.; Taherdoost, H. Barriers and Enablers of AI Adoption in Human Resource Management: A Critical Analysis of Organizational and Technological Factors. Information 2025, 16, 51. [Google Scholar] [CrossRef]
  40. Alsaif, A.; Sabih Aksoy, M. AI-HRM: Artificial Intelligence in Human Resource Management: A Literature Review. J. Comput. Commun. 2023, 2, 1–7. [Google Scholar] [CrossRef]
  41. Sakka, F.; El Maknouzi, M.E.H.; Sadok, H. Human Resource Management in the Era of Artificial Intelligence: Future HR Work Practices, Anticipated Skill Set, Financial and Legal Implications. Acad. Strateg. Manag. J. 2022, 21, 1–14. [Google Scholar]
  42. Islam, M.; Mamun, A.A.; Afrin, S.; Ali Quaosar, G.A.; Uddin, M.A. Technology Adoption and Human Resource Management Practices: The Use of Artificial Intelligence for Recruitment in Bangladesh. South Asian J. Hum. Resour. Manag. 2022, 9, 324–349. [Google Scholar] [CrossRef]
  43. Rani, S. Human Resource Management and Artificial Intelligence. Int. Res. J. Manag. Sociol. Humanit. 2019, 10, 17–25. [Google Scholar]
  44. Abdeldayem, M.M.; Aldulaimi, S.H. Trends and Opportunities of Artificial Intelligence in Human Resource Management: Aspirations for Public Sector in Bahrain. Int. J. Sci. Technol. Res. 2020, 9, 3867–3871. [Google Scholar]
  45. Budhwar, P.; Chowdhury, S.; Wood, G.; Aguinis, H.; Bamber, G.J.; Beltran, J.R.; Boselie, P.; Lee Cooke, F.; Decker, S.; DeNisi, A. Human Resource Management in the Age of Generative Artificial Intelligence: Perspectives and Research Directions on ChatGPT. Hum. Resour. Manag. J. 2023, 33, 606–659. [Google Scholar] [CrossRef]
  46. Kshetri, N. Evolving Uses of Artificial Intelligence in Human Resource Management in Emerging Economies in the Global South: Some Preliminary Evidence. Manag. Res. Rev. 2021, 44, 970–990. [Google Scholar] [CrossRef]
  47. Tewari, I.; Pant, M. Artificial Intelligence Reshaping Human Resource Management: A Review. In Proceedings of the 2020 IEEE International Conference on Advent Trends in Multidisciplinary Research and Innovation (ICATMRI), Buldhana, India, 30 December 2020; IEEE: New York, NY, USA, 2020; pp. 1–4. [Google Scholar]
  48. Sithambaram, R.A.; Tajudeen, F.P. Impact of Artificial Intelligence in Human Resource Management: A Qualitative Study in the Malaysian Context. Asia Pac. J. Hum. Resour. 2023, 61, 821–844. [Google Scholar] [CrossRef]
  49. Evseeva, S.; Evseeva, O.; Burmistrov, A.; Siniavina, M. Application of Artificial Intelligence in Human Resource Management in the Agricultural Sector. Proc. E3S Web Conf. 2021, 258, 01010. [Google Scholar] [CrossRef]
  50. Iqbal, F.M. Can Artificial Intelligence Change the Way in Which Companies Recruit, Train, Develop and Manage Human Resources in Workplace. Asian J. Soc. Sci. Manag. Stud. 2018, 5, 102–104. [Google Scholar] [CrossRef]
  51. Kambur, E. The Effect of Artifical Intelligence on Human Resources Employees. Alanya Akad. Bakış 2021, 5, 1479–1492. [Google Scholar] [CrossRef]
  52. Oztas, B.; Cetinkaya, D.; Adedoyin, F.; Budka, M.; Dogan, H.; Aksu, G. Perspectives from Experts on Developing Transaction Monitoring Methods for Anti-Money Laundering. In Proceedings of the 2023 IEEE International Conference on e-Business Engineering (ICEBE), Sydney, Australia, 4–6 November 2023; IEEE: New York, NY, USA, 2023; pp. 39–46. [Google Scholar]
  53. Parker, S.K.; Ballard, T.; Billinghurst, M.; Collins, C.; Dollard, M.; Griffin, M.A.; Johal, W.; Jorritsma, K.; Kowalkiewicz, M.; Kyndt, E.; et al. Quality Work in the Future: New Directions via a Co-Evolving Sociotechnical Systems Perspective. Aust. J. Manag. 2025. [Google Scholar] [CrossRef]
  54. Bader, V.; Kaiser, S. Algorithmic Decision-Making? The User Interface and Its Role for Human Involvement in Decisions Supported by Artificial Intelligence. Organization 2019, 26, 655–672. [Google Scholar] [CrossRef]
  55. Hristov, V.D.; Saliev, D.N.; Slavov, D.V. Artificial Intelligence Systems for Warehouses Stocks Control. In Proceedings of the 2022 8th International Conference on Energy Efficiency and Agricultural Engineering (EE&AE), Ruse, Bulgaria, 30 June–2 July 2022; IEEE: New York, NY, USA, 2022; pp. 1–5. [Google Scholar]
  56. Autor, D. Polanyi’s Paradox and the Shape of Employment Growth; National Bureau of Economic Research: Cambridge, MA, USA, 2014. [Google Scholar]
  57. Yawalkar, M.V.V. A Study of Artificial Intelligence and Its Role in Human Resource Management. Int. J. Res. Anal. Rev. (IJRAR) 2019, 6, 20–24. [Google Scholar]
  58. Charlwood, A.; Guenole, N. Can HR Adapt to the Paradoxes of Artificial Intelligence? Hum. Resour. Manag. J. 2022, 32, 729–742. [Google Scholar] [CrossRef]
  59. Suseno, Y.; Hudik, M.; Fang, E.S.; Guo, Z. Employee Attitudes, Technological Anxiety, and Change Readiness for Artificial Intelligence Adoption. In Proceedings of the Academy of Management Proceedings, Virtual, 7–11 August 2020; Academy of Management: Briarcliff Manor, NY, USA, 2020; Volume 2020, p. 20045. [Google Scholar]
  60. Soleimani, M.; Arrowsmith, J.; Intezari, A.; Pauleen, D.J. Mitigating Bias in AI-Powered HRM. In Research Handbook on Human Resource Management and Disruptive Technologies; Edward Elgar Publishing: Cheltenham, UK, 2024; pp. 39–50. ISBN 1-80220-924-7. [Google Scholar]
  61. Mukherjee, A.N. Application of Artificial Intelligence: Benefits and Limitations for Human Potential and Labor-Intensive Economy–an Empirical Investigation into Pandemic Ridden Indian Industry. Manag. Matters 2022, 19, 149–166. [Google Scholar] [CrossRef]
  62. Walton, P. Artificial Intelligence and the Limitations of Information. Information 2018, 9, 332. [Google Scholar] [CrossRef]
  63. Franken, S.; Wattenberg, M. The Impact of AI on Employment and Organisation in the Industrial Working Environment of the Future. In Proceedings of the ECIAIR 2019 European Conference on the Impact of Artificial Intelligence and Robotics, Oxford, UK, 31 October–1 November 2019; Academic Conferences and Publishing Limited: Reading, UK, 2019; Volume 31. [Google Scholar]
  64. Zirar, A. Can Artificial Intelligence’s Limitations Drive Innovative Work Behaviour? Rev. Manag. Sci. 2023, 17, 2005–2034. [Google Scholar] [CrossRef]
  65. ÓhÉigeartaigh, S.S.; Whittlestone, J.; Liu, Y.; Zeng, Y.; Liu, Z. Overcoming Barriers to Cross-Cultural Cooperation in AI Ethics and Governance. Philos. Technol. 2020, 33, 571–593. [Google Scholar] [CrossRef]
  66. Koshiyama, A.; Kazim, E.; Treleaven, P.; Rai, P.; Szpruch, L.; Pavey, G.; Ahamat, G.; Leutner, F.; Goebel, R.; Knight, A. Towards Algorithm Auditing: Managing Legal, Ethical and Technological Risks of AI, ML and Associated Algorithms. R. Soc. Open Sci. 2024, 11, 230859. [Google Scholar] [CrossRef] [PubMed]
  67. Rodrigues, R. Legal and Human Rights Issues of AI: Gaps, Challenges and Vulnerabilities. J. Responsible Technol. 2020, 4, 100005. [Google Scholar] [CrossRef]
  68. Stahl, B.C.; Andreou, A.; Brey, P.; Hatzakis, T.; Kirichenko, A.; Macnish, K.; Shaelou, S.L.; Patel, A.; Ryan, M.; Wright, D. Artificial Intelligence for Human Flourishing–Beyond Principles for Machine Learning. J. Bus. Res. 2021, 124, 374–388. [Google Scholar] [CrossRef]
  69. Kunle-Lawanson, N.O. The Role of AI in Information Security Risk Management. World J. Adv. Eng. Technol. Sci. 2022, 7, 308–319. [Google Scholar] [CrossRef]
  70. Ryan, C. Refereeing Articles Including SEM–What Should Referees Look For? Tour. Crit. Pract. Theory 2020, 1, 47–61. [Google Scholar] [CrossRef]
  71. Richter, N.F.; Sinkovics, R.R.; Ringle, C.M.; Schlägel, C. A Critical Look at the Use of SEM in International Business Research. Int. Mark. Rev. 2016, 33, 376–404. [Google Scholar] [CrossRef]
  72. Hair, J.F.; Sarstedt, M.; Ringle, C.M.; Mena, J.A. An Assessment of the Use of Partial Least Squares Structural Equation Modeling in Marketing Research. J. Acad. Mark. Sci. 2012, 40, 414–433. [Google Scholar] [CrossRef]
  73. Nunnally, J.; Bernstein, I. The Assessment of Reliability. Psychometric Theory, 3, 248–292. Nyhan, RC, and HA Marlowe.(1997). Development And Psychometric Properties Of The Organizational Trust Inventory. Eval. Rev. 1994, 21, 614–635. [Google Scholar]
  74. Hair, J.F.; Black, W.C.; Babin, B.J.; Anderson, R.E.; Tatham, R.L. Multivariate Data Analysis, 7th ed.; Pearson: New York, NY, USA, 2010. [Google Scholar]
  75. Fornell, C.; Larcker, D.F. Evaluating Structural Equation Models with Unobservable Variables and Measurement Error. J. Mark. Res. 1981, 18, 39–50. [Google Scholar] [CrossRef]
  76. Fornell, C. A Comparative Analysis of Two Structural Equation Models: LISREL and PLS Applied to Market Data; The University of Michigan: Ann Arbor, MI, USA, 1981. [Google Scholar]
  77. Hair, J.; Alamer, A. Partial Least Squares Structural Equation Modeling (PLS-SEM) in Second Language and Education Research: Guidelines Using an Applied Example. Res. Methods Appl. Linguist. 2022, 1, 100027. [Google Scholar] [CrossRef]
  78. Roemer, E.; Schuberth, F.; Henseler, J. HTMT2–an Improved Criterion for Assessing Discriminant Validity in Structural Equation Modeling. Ind. Manag. Data Syst. 2021, 121, 2637–2650. [Google Scholar] [CrossRef]
  79. Radomir, L.; Moisescu, O.I. Discriminant Validity of the Customer-Based Corporate Reputation Scale: Some Causes for Concern. J. Prod. Brand Manag. 2020, 29, 457–469. [Google Scholar] [CrossRef]
  80. Ab Hamid, M.R.; Sami, W.; Sidek, M.M. Discriminant Validity Assessment: Use of Fornell & Larcker Criterion versus HTMT Criterion. J. Phys. Conf. Ser. 2017, 890, 012163. [Google Scholar]
  81. Hu, L.; Bentler, P.M. Fit Indices in Covariance Structure Modeling: Sensitivity to Underparameterized Model Misspecification. Psychol. Methods 1998, 3, 424. [Google Scholar] [CrossRef]
  82. Ullman, J.B.; Bentler, P.M. Structural Equation Modeling. In Handbook of Psychology, 2nd ed.; John Wiley & Sons, Inc.: Hoboken, NJ, USA, 2012; Volume 2. [Google Scholar]
Sustainability 17 05815 g001
Figure 1. Distribution of innovation adoption. Source: Dearing and Cox (2018) [30].
Figure 1. Distribution of innovation adoption. Source: Dearing and Cox (2018) [30].
Sustainability 17 05815 g001
Sustainability 17 05815 g002
Figure 2. Research model. Source: authors’ own work.
Figure 2. Research model. Source: authors’ own work.
Sustainability 17 05815 g002
Sustainability 17 05815 g003
Figure 3. Key human rights issues in AI. Source: Key human rights issues in AI (Stahl et al. 2021) [68].
Figure 3. Key human rights issues in AI. Source: Key human rights issues in AI (Stahl et al. 2021) [68].
Sustainability 17 05815 g003
Sustainability 17 05815 g004
Figure 4. Correlation heatmap. Source: authors’ own work. Note: PAITx, PAIRSx, PAIPAx, PAICRx, and PAICDx are items in the training, recruitment and selection, performance appraisal, compensation and rewards, and career development dimensions.
Figure 4. Correlation heatmap. Source: authors’ own work. Note: PAITx, PAIRSx, PAIPAx, PAICRx, and PAICDx are items in the training, recruitment and selection, performance appraisal, compensation and rewards, and career development dimensions.
Sustainability 17 05815 g004
Sustainability 17 05815 g005
Figure 5. Measurement model. Source: authors’ own work.
Figure 5. Measurement model. Source: authors’ own work.
Sustainability 17 05815 g005
Table 1. AI applications in HRM.
Table 1. AI applications in HRM.
StudyMain AI Applications in HRM
[44]Training and development, compensation, career path, performance analysis, recruitment, and talent acquisition.
[45]Recruitment, talent management, HR shared service, learning and development, reporting and analytics, and self-service assistance.
[25]Recruitment and selection process, planned training and development process, and tactical performance appraisal.
[46]Recruitment and selection, and the development, retainment, and productive utilization of employees.
[4]Recruitment till performance appraisal of employees.
[47]Recruitment and selection, compensation, performance management, employee retention, training and development, and employee engagement.
[48]Recruitment, talent management, HR shared service operations, learning and development, reporting and analytics, and self-service assistance.
[41]Talent acquisition and management, performance appraisal, training and development, employee motivation and engagement.
[49]Recruitment, onboarding, assessment, training and development, talent management, salary system, culture and engagement management, and management and leadership.
[50]Recruitment, training, development, and retention of workers in organizations.
[5]Recruitment: filtering, preselection, performance assessment, and measurement of performance data management, pay, and rewards.
[42]Recruitment: adoption and use of AI.
[20]Recruitment, development, and performance appraisal.
AI applications in HRM have been classified into six scenarios: A, B, C, and D are directly related to recruitment and development, and scenarios E and F are related to performance appraisal.
[43]Recruitment.
[13]Recruitment process: screening and interview process, reducing administrative burden, selection process, reducing discrimination, increasing efficiency, and enriching learning.
Table 2. Sample’s characteristics (N = 234).
Table 2. Sample’s characteristics (N = 234).
CharacteristicFrequencyPercentage
Gender
  Male18679.5
  Female4820.5
Age
  Less than 30--
  30 to 347532.1
  35 to 393314.1
  40 to 446025.6
  45 to 493012.8
  50 and above3615.4
Educational Level
  High diploma187.8
  University graduates13256.4
  High school graduates187.7
  Post-graduate6628.2
Total Working Time
  0–5 years3916.8
  6–10 years7833.3
  11–15 years6025.6
  16–20 years2410.3
  More than 21 years3314.1
Total Working Time at Current Organization
  0–5 years7833.3
  6–10 years10243.6
  11–15 years3012.8
  16–20 years31.3
  More than 21 years219.0
Total Working Time in HRM Department
  0–5 years9942.3
  6–10 years6929.5
  11–15 years4519.2
  16–20 years156.4
  More than 21 years62.6
Table 3. AI knowledge and frequency of use.
Table 3. AI knowledge and frequency of use.
FrequencyPercentage
AI Knowledge
  Very little2410.3
  Little6929.5
  Average10846.2
  Much2410.3
  Very much93.9
AI Skills
  Very little4519.2
  Little8435.9
  Average7833.3
  Much187.8
  Very much93.9
Frequency of AI Use
  Very little4820.2
  Little7230.8
  Average5121.8
  Much5423.1
  Very much93.9
Frequency of AI Use in HRM
  Very little9641.0
  Little6025.6
  Average4820.5
  Much219.0
  Very much93.9
Table 4. Descriptive statistics of compensation and rewards.
Table 4. Descriptive statistics of compensation and rewards.
ItemMean (SD)MedianSkewnessKurtosis
I think it will be easier to adapt to the possible changes in the salary system (time, per piece, premium) with artificial intelligence.3.65 (0.97)3−0.02−0.70
I think that artificial intelligence technology can hinder to delay due to human reasons in salary, premium, prize, bonus, and such payments.3.47 (1.00)4−0.20−0.12
I think that my extra wages (premium, bonus, overtime) except for my salary will be calculated correctly with artificial intelligence technology.3.62 (1.09)4−0.33−0.75
I think that artificial intelligence technology will help to determine the amount of salary I will receive fairly.3.42 (1.12)3−0.11−0.80
PAICR3.54 (0.78)3.50−0.020.10
Table 5. Descriptive statistics of career development.
Table 5. Descriptive statistics of career development.
ItemMean (SD)MedianSkewnessKurtosis
I think that artificial intelligence technology will help me which qualifications I should have in achieving my dream career.4 (0.95)4−0.730.05
I think that artificial intelligence technology will automate the wage rise depending on the skill increase.3.73 (0.90)4−0.08−0.89
I think that artificial intelligence technology will make it easier to recognize the employees who really deserve promotion in their career.3.58 (1.12)4−0.17−1.13
I think that the appropriate team member can be identified quickly via artificial intelligence technology.3.64 (1.10)4−0.30−0.79
I think that artificial intelligence technology will help me to acquire the necessary skills for my career plan.3.76 (0.98)4−0.32−0.88
I think that artificial intelligence technology will help me to determine my ideal career plan.3.60 (0.99)4−0.01−1.07
PAICD3.72 (0.80)3.83−0.09−0.81
Table 6. Descriptive statistics of performance appraisal.
Table 6. Descriptive statistics of performance appraisal.
ItemMean (SD)MedianSkewnessKurtosis
I think that over-competitive performance measurement with artificial intelligence technology will be obsolete.3.32 (0.96)30.12−0.19
I think that artificial intelligence technology can accurately predict how the staff will perform in the future.3.29 (1.05)3−0.28−0.02
I think that I will not lose motivation when my performance is measured with artificial intelligence technology.3.37 (1.11)3−0.10−0.62
I think that artificial intelligence technology will determine the performance determination criteria correctly.3.46 (1.02)3−0.15−0.57
When my performance is measured with artificial intelligence technology, I think this will positively affect the corporate culture.3.44 (1.08)3−0.26−0.43
When my performance is measured with artificial intelligence technology, I think this will have a positive effect on the success of the company.3.56 (0.97)3−0.22−0.21
PAIPA3.41 (0.80)3.33−0.07−0.04
Table 7. Descriptive statistics of recruitment and selection.
Table 7. Descriptive statistics of recruitment and selection.
ItemMean (SD)MedianSkewnessKurtosis
I think that artificial intelligence technology will reduce the time spent on finding candidates.3.83 (1.02)4−0.69−0.01
I think that artificial intelligence technology will save the monotony of the job done during the process of finding candidates.3.58 (0.96)4−0.31−0.09
I think that artificial intelligence technology will gain access to more qualified candidates.3.78 (0.89)4−0.18−0.69
I think that artificial intelligence technology will reduce the stress of finding the appropriate candidate.3.78 (0.86)4−0.18−0.69
I think that the candidate resume will be able to examine in detail with artificial intelligence technology.3.74 (0.94)4−0.590.32
I think that the most suitable personnel will be selected through artificial intelligence technology.3.46 (1.01)3−0.27−0.22
I think the technology of artificial intelligence will reduce the time spent selecting personnel.3.76 (0.90)4−0.470.05
I think that artificial intelligence technology will prevent the instability of choosing personnel.3.59 (0.93)40.03−0.88
I think that my professional knowledge will be kept up to date with in-company training courses through artificial intelligence technology.3.78 (0.83)4−0.530.59
PAIRS3.70 (0.68)3.670.24−0.89
Table 8. Descriptive statistics of training.
Table 8. Descriptive statistics of training.
ItemMean (SD)MedianSkewnessKurtosis
I believe that in-company training courses via artificial intelligence technology will lead to a successful training program.3.59 (1.01)4−0.40−0.34
I think that when the in-company training courses take place with artificial intelligence technology, the restrictions regarding the place where the training will be given will be.3.50 (0.95)3−0.23−0.10
I think that artificial intelligence technology will increase accessibility to in-company training courses.3.77 (0.86)4−0.380.15
I think that artificial intelligence technology will reduce the attention deficit that I experienced in classical in-company training courses processes.3.60 (0.82)30.020.13
I think that artificial intelligence technology will reduce the time spent on in-company training.3.69 (0.93)4−0.33−0.26
PAIT3.63 (0.75)3.60−0.270.73
Table 9. Comparison of dimension means across demographic characteristics.
Table 9. Comparison of dimension means across demographic characteristics.
DimensionAgeEducationTotal Working Time
F-ValuepF-ValuepF-Valuep
PAICRF (4, 229) = 2.270.063F (3, 230) = 6.840.000F (4, 229) = 2.320.058
PAICDF (4, 229) = 0.190.432F (3, 230) = 1.490.217F (4, 229) = 3.550.007
PAIPAF (4, 229) = 0.76,0.555F (3, 230) = 2.360.072F (4, 229) = 4.870.001
PAIRSF (4, 229) = 2.630.035F (3, 230) = 8.340.000F (4, 229) = 0.410.796
PAITF (4, 229) = 0.500.718F (3, 230) = 4.210.007F (4, 229) = 0.490.737
Table 10. Model construct reliability and convergent validity.
Table 10. Model construct reliability and convergent validity.
ConstructItemFactor LoadingCronbach’s AlphaCRAVE
PAICDPAICD10.710.740.890.57
PAICD20.77
PAICD30.70
PAICD40.76
PAICD50.79
PAICD60.80
PAICRPAICR10.620.880.750.44
PAICR20.50
PAICR30.67
PAICR40.81
PAIPAPAIPA10.500.870.880.55
PAIPA20.63
PAIPA30.77
PAIPA40.83
PAIPA50.77
PAIPA60.87
PAIRSPAIRS10.560.890.890.48
PAIRS20.69
PAIRS30.71
PAIRS40.66
PAIRS50.69
PAIRS60.69
PAIRS70.79
PAIRS80.71
PAIRS90.73
PAITPAIT10.740.880.880.60
PAIT20.80
PAIT30.76
PAIT40.74
PAIT50.83
Knowledge and useAI knowledge0.780.880.880.66
AI skills0.86
Frequency of AI use in HRM0.70
Frequency of AI use0.89
Table 11. Model fit metrics.
Table 11. Model fit metrics.
MetricValue
CFI0.84
TLI0.80
RMSEA0.07
SRMR0.08
Table 12. Fornell–Larcker criterion.
Table 12. Fornell–Larcker criterion.
ConstructPAICDPAICRPAIPAPAIRSPAITKnowledge and Use
PAICD0.75
PAICR0.760.66
PAIPA0.850.760.74
PAIRS0.710.590.660.69
PAIT0.680.590.510.720.77
Knowledge and use0.230.050.250.150.200.81
Table 13. HTMT ratio.
Table 13. HTMT ratio.
ConstructPAICDPAICRPAIPAPAIRSPAITKnowledge and Skills
PAICD1
PAICR0.771
PAIPA0.880.811
PAIRS0.710.630.701
PAIT0.670.630.540.731
Knowledge and skills0.310.090.280.160.221
Table 14. Path analysis.
Table 14. Path analysis.
Pathβ CoefficientStd. Errorp-Value
PAICD—knowledge and use0.280.070.000
PAICR—knowledge and use0.040.070.558
PAIPA—knowledge and use0.170.050.001
PAIRS—knowledge and use0.120.060.042
PAIT—knowledge and use0.210.080.006
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Alshahrani, S.T.; Choukir, J.; Albelali, S.; AlShalhoob, A.A. Perceptions of the Impact of AI on Human Resource Management Practices Among Human Resource Managers Working in the Chemical Industry in Saudi Arabia. Sustainability 2025, 17, 5815. https://doi.org/10.3390/su17135815

AMA Style

Alshahrani ST, Choukir J, Albelali S, AlShalhoob AA. Perceptions of the Impact of AI on Human Resource Management Practices Among Human Resource Managers Working in the Chemical Industry in Saudi Arabia. Sustainability. 2025; 17(13):5815. https://doi.org/10.3390/su17135815

Chicago/Turabian Style

Alshahrani, Saeed Turki, Jamel Choukir, Saja Albelali, and Abdulaziz Abdulmohsen AlShalhoob. 2025. "Perceptions of the Impact of AI on Human Resource Management Practices Among Human Resource Managers Working in the Chemical Industry in Saudi Arabia" Sustainability 17, no. 13: 5815. https://doi.org/10.3390/su17135815

APA Style

Alshahrani, S. T., Choukir, J., Albelali, S., & AlShalhoob, A. A. (2025). Perceptions of the Impact of AI on Human Resource Management Practices Among Human Resource Managers Working in the Chemical Industry in Saudi Arabia. Sustainability, 17(13), 5815. https://doi.org/10.3390/su17135815

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop