The Smart Shift: A Knowledge Management and Industrial–Organizational Psychology Perspective on Digital Transformation and Sustainable Well-Being Among SMEs

Abstract

Artificial Intelligence (AI) has become a fundamental driver of digital transformation, reshaping organizational management, leadership behavior, and the sustainability of human work systems. Despite its potential to improve performance, few studies have explored how executives psychologically respond to AI awareness and its implications for sustainable well-being. Drawing upon Knowledge Management (KM) theory and Industrial–Organizational (I–O) Psychology, this study examines how senior executives’ awareness of AI (AIA) affects job burnout, with job insecurity serving as a mediator and self-esteem as a moderator. Data were collected from 615 CEOs and senior managers of small and medium-sized enterprises (SMEs) in the United Arab Emirates (UAE) and analyzed using structural equation modeling (Smart PLS 4). The results reveal that higher AI awareness intensifies burnout primarily through increased perceptions of job insecurity; however, executives with higher self-esteem demonstrate resilience to these effects. By framing AIA within the Knowledge Management (KM) theory, this study contributes to the existing KM literature by revealing how organizations create, maintain, and use knowledge assets in the digital transformation environment. Our findings underscore the necessity for organizations to set up innovative initiatives, flexible organizational structures, targeted training, and mental health support while adopting AI technologies. Overall, this study highlights the critical intersection between digital Knowledge Management and the mental health of executives, aligning with Sustainable Development Goal 3 (Good Health and Well-Being).

1. Introduction

In recent years, Artificial Intelligence (AI) technologies have evolved rapidly and are now embedded across various industries, enabling organizations to generate substantial economic and strategic value [1]. Application of hybrid systems, particularly AI, is the entrance door of organizational innovation and exploitation of knowledge sources [2]. AI is considered a key technology for the Fourth Industrial Revolution as it reconfigures production, decision-making, and service delivery across industries and leads companies and countries to compete based on digital capabilities rather than only traditional assets [3]. AI is reconfiguring the global governance landscape by providing new possibilities for improving systemic resilience while also raising significant ethical, social, and geopolitical dilemmas [4]. AI is becoming more and more powerful and pervasive in society and presents cross-border risks, such as loss of jobs, security, bias, and environmental damage, leading to a moral need for international governance [5].

Top management must create AI knowledge and awareness among its employees and educate them about the capabilities and value of AI, as well as the limitations associated with digital transformation [6]. AI is particularly central to the ambitions of digitally progressive economies, such as the United Arab Emirates (UAE), where it supports national strategies for sustainable growth, competitiveness, and technological self-sufficiency [7]. The UAE’s national agenda positions AI as a key driver for achieving the United Nations Sustainable Development Goals (SDGs), particularly those related to innovation, economic diversification, and well-being [8,9,10].

In this policy backdrop, small and medium-sized enterprises (SMEs) are the backbone of the UAE’s diversification strategy; however, little is known about how awareness of AI in senior executives influences psychological and organizational outcomes.

In addition to its technological aspect, AI adoption also correlates with the ESG responsibilities of firms in how they disclose sustainability information and contribute to performance outcomes [11,12]. Good ESG practices enable ethical governance, the well-being of employees, and transparency, which are desirable aspects for long-term competition and stakeholder faith [13]. In economies such as the UAE, integrating AI into governance and social systems can enhance ESG performance by improving decision-making, accountability, and the traceability of sustainability disclosures [14]. However, few studies have examined how leadership-level digital awareness, particularly executives’ AI awareness (AIA), translates technological advancement into socially responsible and psychologically sustainable behavior.

Knowledge is now the main factor of production rather than financial capital or machinery [15]. It is the key element of competitive advantage for employees and organizations in the labor and service market [16]. In the era in which AI technologies are becoming more prevalent in the workplace, it is essential to comprehend how these technologies and to what extent affect employees’ attitudes and actions [17]. Previous studies have investigated the effect of AIA on job instability [12,18,19,20]. Workers who are exposed to AI technology (either physically or psychologically) would report greater job insecurity [21]. With the rise of AI in the workplace, AI awareness and the employees’ perception of their job replacement by AI are directly related to the employees’ emotional status, feelings, and work–life balance [22]. Recent empirical work further indicates that the psychosocial consequences of digital transformation are intensifying. Rapid technological change generates new stressors, such as technostress, ambiguous work–life boundaries, and a growing sense of digital overload, all of which can undermine productivity and well-being [23,24]. In a study by (Zeike et al., 2019) [25] managers who were evaluated were found to have poor well-being in 25.4% of cases, suggesting that they are vulnerable to depression. A total of 10.3% of managers in the same survey were previously acknowledged as having depressive symptoms. Since leaders directly influence organizational culture and employee morale [26], their own mental health and sense of stability are vital for sustaining firm-level performance and social responsibility [27]. While prior studies have explored the effects of AI on psychological and behavioral outcomes such as anxiety, depression, and turnover [1,18,28,29], few have examined the mediating mechanisms of job insecurity or the protective role of personal resources such as self-esteem, particularly at the leadership level. This study addresses these shortcomings by examining how executives’ AI awareness influences job burnout, job insecurity, and self-esteem within the broader context of ESG and executive governance.

Moreover, the profitability of a business in the modern business environment largely depends upon its capacity for learning and adaptation. When businesses employ dynamic procedures that foster, leverage, and inspire individuals to enhance and share their knowledge, learning ability will grow. Such procedures embody the concept of Knowledge Management (KM) [15]. However, ref. [30] found that knowledge responsiveness, the ability to react quickly to new knowledge signals, can negatively affect business intelligence system performance, even while contributing positively to sustainability outcomes, underscoring that KM adaptation entails potential knowledge–risk trade-offs. AI redefines these social learning dynamics by mediating how individuals participate in knowledge communities or “Ba,” transforming collaboration and collective sensemaking into hybrid human–machine interactions [31]. Additionally, investments in psychological capital development boost Knowledge Management, enhance digital communication, and sustain employees’ well-being [32]. The adoption of AI offers advantages such as increased productivity, competencies, and innovation while simultaneously presenting challenges, including employee resistance, cultural mismatch, ethical issues, and leadership challenges. To overcome these challenges and ensure the successful use of AI, effective leadership, skill promotion, and transparent communication are required [33].

From a theoretical perspective, this study, along with KM theory (Nonaka, 2009) [34], draws on stakeholder theory [35] to explain how executives’ AI awareness connects digital transformation with sustainable corporate outcomes. Stakeholder theory suggests that leaders’ well-being and responsiveness to stakeholder expectations shape how effectively they manage technological disruption. Recent evidence shows that internal and external stakeholder support enhances the positive impact on technological adoption in organizational success [36]. To acknowledge this, these perspectives highlight that executives’ cognitive and psychological engagement with AI is not only a technological issue but also a strategic mechanism through which firms meet stakeholder expectations and strengthen ESG performance.

2. Literature Review

To enhance conceptual rigor, this study synthesizes Knowledge Management (KM) and Industrial–Organizational (I–O) Psychology with stakeholder theory. The theoretical frameworks explain how executives’ psychological states during AI adoption influence not only their personal well-being but also stakeholder trust and sustainability performance [37]. In addition, the Job Demands–Resources (JD–R) model is applied to capture the psychological strain mechanisms linking AI awareness, job insecurity, and burnout [38]. This integration provides a comprehensive view of how cognitive, organizational, and contextual factors jointly shape executives’ experiences in AI-driven environments. Evidence from SME contexts further shows that business intelligence systems stimulate entrepreneurial leadership, which in turn enhances organizational sustainability [30,39]. This highlights the cognitive and leadership roles of executives as central to sensemaking in digitally transforming SMEs. According to the knowledge-based theory, knowledge is seen as the most important strategic resource [34], which is generated, shared, and effectively used to support strategic objectives. Implementing AI in the organization is considered a double-edged sword from the perspective of Knowledge Management. AI awareness improves Knowledge Management by speeding up and automating the regular organizational tasks; on the other hand, increasing AI awareness may raise concerns among employees and make them feel insecure, as their jobs are being automated [40]. When employees feel that their knowledge and skills are being rendered obsolete by AI, they may experience a sense of loss and be less motivated, contributing to burnout [18]. According to KM theory, knowledge is a strategic resource that enables competitive advantage through creation, sharing, and utilization (Nonaka, 2009) [34]. As organizations engage with AI, this process is accelerated. Generative AI and advanced analytics automate data-driven decision-making and can magnify the benefits of Knowledge Management [40]. Recent work confirms that AI accelerates the SECI spiral of knowledge conversion by codifying tacit insights into explicit, machine-processable forms, thereby shifting human roles toward validation and oversight rather than generation [31].

At the same time, AI awareness introduces disruption into existing knowledge hierarchies. When employees or executives perceive that their expertise may become obsolete, it generates tension between innovation and obsolescence [41]. In sustainability-oriented firms, effective Knowledge Management hinges on leadership’s ability to transform technological awareness into shared learning and collective competencies, thereby enhancing both productivity and the integrity of ESG reporting. However, when AI awareness is perceived as threatening, by signaling potential job displacement, it can give rise to job insecurity and psychological strain, undermining knowledge sharing and organizational trust [17].

From a psychological standpoint, AIA may be a double-edged sword: while it increases efficiency and decision accuracy, it can also create anxiety and insecurity as employees or executives perceive their expertise becoming obsolete [18]. When individuals feel that technology threatens the relevance of their knowledge, they experience a loss of self-efficacy and motivation, which may result in emotional exhaustion and burnout [41].

From a stakeholder theory perspective, executives have an ethical duty to balance technological efficiency with employee welfare, as neglecting the human dimension of AI transformation may erode stakeholder confidence [35,42]. Empirical evidence suggests that internal and external stakeholder support enhances the positive impact of technology adoption on organizational success [36]. In this regard, ref. [43] argues that ESG rating divergence acts as an informal institutional pressure, compelling firms to align through transparent and digitalized disclosure practices. ESG performance and sustainability disclosure thus function as visible mechanisms through which organizations signal their legitimacy [44]. By demonstrating responsible and transparent behaviors, they meet societal norms and bolster stakeholder trust [45]. In such contexts, executives’ psychological capacity to manage AI-driven change, as explained by I–O psychology and JD–R theory, becomes a prerequisite for maintaining stakeholder trust, since these internal capabilities determine how effectively the firm responds to external demands for responsible, transparent digital, and ESG behaviors.

2.1. Theoretical and Hypothesis Development

2.1.1. AI Awareness and Job Burnout

In the era of the fast spread of AI, employees are becoming more knowledgeable about and utilizing AI. In this environment, employees’ behaviors are mostly affected by their knowledge [41]. In spite of the fact that AI triggers some challenges, several organizations are encouraging or mandating employees to integrate AI into their daily tasks [46]. However, the progression of AI in the workplace might lead to burnout [18]. Burnout is a behavioral reaction to workplace challenges manifested in three categories: (1) exhaustion, (2) depersonalization disorder, and (3) lowered competency and low achievement [47]. Moreover, a study by Iram et al. discovered that employee AIA facilitates the association between knowledge hiding and technological instability. Adopting proper leadership might reduce knowledge hiding by manager–employee collaboration and encouraging innovative initiatives [48]. Additionally, awareness of AI ultimately shapes the performance of the employees [46]. However, in the educational context, AIA has a significant positive influence on knowledge sharing but demonstrated a significant negative impact on the performance of teachers [49]. Furthermore, the study findings demonstrate that AIA possesses a strong relationship with interpersonal conflicts, turnover rate, and counterproductive behavior [50]. Organizational support is necessary to reduce counterproductive work behavior and turnover intention, creating a work environment based on honesty, trust, fairness, and respect [51]. Previous studies have also unveiled that employees’ AIA hurts competitive productivity, psychological safety, work engagement, organizational commitment, and satisfaction [52] and substantially affects turnover rate, depression, job overload, and burnout [12,53]. However, the extent to which AI awareness leads to burnout depends on contextual factors, such as organizational support and communication transparency, which shape whether AI is seen as a challenge or a threat [51]. Stakeholder theory posits that organizations must balance technological advancement with the social and psychological well-being of their internal stakeholders to maintain sustainable legitimacy and trust [36,42]. Employees with high knowledge of AI may believe that their job is considerably unreliable; they worry that digital types of machinery could replace their positions one day [18]. This perception of technological threat erodes employees’ sense of control and competence, depleting psychological resources that ultimately manifest as burnout [46,47]. Employees’ awareness of AI is equivalent to hampering their ability to perform their jobs [54]. Therefore, based on the literature, the following hypothesis is proposed:

H1. 

AI awareness has a positive influence on job burnout.

2.1.2. Job Insecurity as a Mediator

Stakeholder theory clarifies the mediating mechanism of job insecurity between AI awareness and burnout. Stakeholder theory explains that when organizations fail to support their internal stakeholders through transparent communication and skill development, employees’ perceived job insecurity increases [42].

AI use could impact employees’ professional development and potentially replace their role inside the organization [55]. However, despite robots occasionally being more productive than humans, workers may find it challenging to achieve personal goals. Furthermore, Li et al. [56] proposed that employees’ AIA increases their intent to leave since the excessive use of AI makes them feel insecure and anxious. Because high turnover intentions negatively impact one’s mental health, they are also a strong predictor of job burnout [57]. Thus, awareness of AI initiates cognitive appraisals of job threat, which heightens job insecurity, leading to emotional exhaustion and burnout through the depletion of personal resources [18,46]. The implementation of digital transformation in the workplace will change job structures and management systems, which could make workers dissatisfied, depressed, skeptical, and probably provoke burnout in existing positions [18]. Additionally, awareness of AI has a detrimental effect on workers’ mental health, for instance, increasing their sense of job insecurity [58], triggering employee job burnout [18] and depression [59], and reducing employees’ job competency. Consequently, work instability as a threat to one’s self-efficacy and job fit may be a significant mediator between depression and AI awareness [1]. According to the JD-R theory, workers preserve their assets to prevent losses. Employee insecurity and uncertainty lead to knowledge hiding if they believe that technological advancements like AI are inappropriate or against their interests. Knowledge hiding behavior is deliberately retaining information from colleagues when they inquire [48]. Generally, employees will seek information on AI if they perceive it as a challenge to their jobs, resulting in constructive behavior, like informal learning. However, awareness of AI may be viewed as a danger or obstacle that results in employment instability [60]. Some employees comprehend AIA as a challenge that inspires them to develop and promote their proficiencies [54]. Employees are more inclined to leave their jobs when new technology causes them to feel unsafe and insecure at work [51]. It is important to differentiate human learning, social, exploratory, and meaning-oriented, from machine learning, which is procedural and data-driven [31]. When executives become aware that AI encroaches on exploratory and creative domains, they may experience reduced autonomy and emotional exhaustion, reinforcing the mediating role of job insecurity [61,62,63,64]. This mediating process may be stronger in institutional environments with rapid technological change or weak employee support systems, where uncertainty about job roles is amplified [17,65]. Accordingly, we hypothesized the following:

H2. 

Job insecurity mediates the relationship between AI awareness and job burnout.

2.1.3. Self-Esteem as a Moderator

Drawing from stakeholder theory and Knowledge Management theory, self-esteem can be viewed as a critical personal resource that shapes how individuals interpret AI awareness. Stakeholder theory holds that employee psychological well-being is a fundamental dimension of corporate responsibility and sustainability, influencing how stakeholders evaluate an organization’s legitimacy [13,36]. High self-esteem reflects employees’ psychological capital, allowing them to perceive AI as a collaborative and empowering tool rather than a threat [13,31,36]. Conversely, lower self-esteem intensifies perceived job insecurity, as workers doubt their adaptability to new knowledge systems [1,66,67]. Because self-esteem shapes individuals’ threat appraisal and coping ability, it determines whether AI awareness is processed as a developmental challenge or a resource-depleting threat [67,68].

Self-esteem is a constant psychological attribute that describes a person’s psychological perspective toward their value and capabilities [69,70]. Employees are very watchful to make sure that they are treated properly at work and that their expectations are fulfilled in their exchange relationship with the business in order to preserve their good self-construction [71]. According to psychological theory and the self-esteem construct, workers with high self-esteem are more definitely to have greater prospects for their company when they sign the psychological contract [72]. With the implementation of AI, employees may face a discrepancy between their knowledge of AI and the proficiencies required for new duties and their skills and knowledge, which negatively affects their perspective and diminishes their conceptions of themselves, such as self-esteem. Self-esteem is a personal resource. Thus, the use of AI will result in a decline in self-esteem [1].

The existing literature posits that workers with lower self-esteem are more prone to engage with social media and other applications [73]. Therefore, we assume that those with low self-esteem are more prone to problematic chatbot use (PACU) [74]. Conversely, high self-esteem was associated with lower AIC use and the need to interact with someone [75]. While there is an association between high self-esteem and a preference for social engagement [75], it seems that high self-esteem is mostly associated with positive face-to-face relationships, while low self-esteem may lead to a preference for online social connections [60]. High levels of job instability seemed to predict low self-esteem subsequently, but simultaneously, and to the same extent, low self-esteem will also predict successive high job insecurity [66]. Collaborating with robots may lead employees to face the risk of low self-esteem, which in turn triggers employee burnout [67]. From a knowledge-management standpoint, the collaborative intelligence view of AI suggests that technology can serve as a cognitive partner rather than a substitute; thus, executives with higher self-esteem are more likely to interpret AI awareness as empowerment, while those with lower self-esteem perceive displacement [31]. Moreover, self-esteem threat mediates the indirect impact of employee collaboration with robots on burnout [67]. Prior studies have already exhibited that self-efficacy, self-esteem, and self-control are essential components for achieving a flexible and healthy use of technology [60]. The conceptual model illustrates the effect of AIA on job burnout and the indirect effect of job insecurity, with self-esteem moderating the effect of AIA on job insecurity (see Figure 1). Therefore, we hypothesized the following:

H3. 

Self-esteem moderates the relationship between AI awareness and job insecurity.

3. Materials and Methods

3.1. Sample Selection and Data Sources

The registered number of SME businesses in the United Arab Emirates is 557,000 as of mid-2022. SMEs make up as much as 63.5 percent of the GDP in non-oil industries [76]. A business is categorized as a small enterprise if it employs between 6 and 50 people and a medium enterprise if it employs between 51 and 200 people [77] (see

Table 1. Specification of variables.

Variable	Definition
Dependent and independent variables
Burnout
AI awareness	A four-item scale was modified to test AI awareness, taken from Brougham, D.; Haar, J. 2018 [59].
Job insecurity	A six-item scale was modified to measure job burnout, taken from Brougham, D.; Haar, J. 2018 [59].
Self-esteem	Five items were modified to measure self-esteem, taken from Joseph Ciarrochi and Linda Bilich, 2006 [79].
Control variables
SMEs	Small enterprises, with 6 to 50 employees, or annual revenues that are not more than AED 50 million. Medium enterprises, with 51 to 200 employees, or annual revenues are not more than AED 250 million; United Arab Emirates cabinet, 2016.

). Using SurveyMonkey (SurveyMonkey Inc., San Mateo, CA, USA; www.surveymonkey), the researcher developed an online survey tool. Our study’s population approximates 1,114,000 when a CEO and one senior manager are taken into account for each SME in the United Arab Emirates. The paper clearly outlines the researcher’s use of a quantitative research methodology to examine the information provided by SME CEOs and other senior managers in the United Arab Emirates. The sample size required for the investigation was determined using G*power 3.1 based on the criteria related to the current model [78]. The criteria modified for the study were an effect size (f2 = 0.03), a target statistical power (1 − β = 0.95), and an alpha level of (α = 0.05). According to the software suggestion, the survey should be administered to at least 518 people. Seven hundred questionnaires were distributed via SurveyMonkey. To prevent misunderstandings and conduct pilot research before extensive data collection, 30 managers were eliminated from the study. Data screening for cleaning through IBM SPSS statistics, Version 25 was then performed through five stages: unengaged cases, missing data, duplicate cases, outliers, and normality tests. Lastly, 16 unengaged cases, 3 missing data, and 1 duplicate datum were identified, and there was no outlier based on the box plot graph. Following data screening, 615 data points were deemed suitable for further analysis. Accordingly, the data was also determined to be normal, and the reliability of Cronbach’s alpha was regarded as acceptable for four latent variables (see Table 1).

3.2. Study Measurement

The study’s questionnaire was segmented into discrete sections that were pertinent to the primary components and were created using the scales from earlier research. Four items modified from a [59] scale were used to test AI awareness. The sample item, “As AI replaces workers in my industry, I am concerned about my future,” is illustrative. The measurement had a Cronbach’s alpha of 0.839. Five questions that were modified from the Koo, Curtis, and Ryan [80] scale were used to measure job insecurity. The sample question, “Do you believe you can stop unfavorable things from happening to your working environment?” is expressive. For the measure, 0.816 was the Cronbach’s alpha. Six items that were modified from the Brougham and Haar [59] scale were used to measure job burnout: “My job tasks occasionally make me feel ill.” For the measure, Cronbach’s alpha was 0.839. Lastly, five items that were modified from a scale created by Joseph Ciarrochi and Linda Bilich [79] were used to measure self-esteem. The phrase, “On the whole, I am satisfied with myself,” is an example of a representative item. The Cronbach’s alpha for the variable was 0.830. The survey applied a Likert scale, where 1 represented “strongly disagree” and 5 signified “strongly agree.”

Table 2. Scales, outer loadings (OLs), Cronbach’s alpha, CR, and AVE.

Variable		Scales	OL	Cronbach’s Alpha	CR (rho_a)	CR (rho_c)	Average Variance Extracted (AVE)
AI Awareness	AI1	I am worried about my upcoming years as AI replaces workers in my industry.	0.717	0.839	0.847	0.891	0.672
	AI2	Since AI is replacing workers, I am truly concerned about my future at my job.	0.869
	AI3	And I am concerned that AI may be able to replace the work I do now.	0.857
	AI4	AI might eventually replace me in my line of work.	0.827
Job Insecurity	JI1	Do you believe that you lack the authority to influence changes that could have an impact on your employment at your company?	0.675	0.816	0.814	0.878	0.642
	JI2	Do you believe you can keep unfavorable things from harming your working environment?	0.784
	JI3	Do you believe your company is strong enough to manage issues that impact you?	0.774
	JI4	Do you think your organization is capable of handling problems that affect you?	0.763
	JI5	In your opinion, AI in your industry is a friend	0.743
Job Burnout	JB1	My job tasks occasionally make me feel ill.	0.726	0.839	0.852	0.883	0.558
	JB2	Compared to the past, I usually need more time to unwind and feel better after work.	0.736
	JB3	I frequently feel emotionally drained while working.	0.759
	JB4	I find myself talking negatively about my work more and more frequently.	0.791
	JB5	These days, I work nearly entirely mechanically at work and tend to think less.	0.722
	JB6	Usually, I feel exhausted and worn out after work.	0.747
Self-Esteem	SE1	I am generally happy with myself.	0.666	0.830	0.864	0.882	0.653
	SE2	I believe I possess several positive traits.	0.831
	SE3	I can perform tasks just as well as the majority of people.	0.849
	SE4	I believe that I am A valuable individual.	0.716
	SE5	I have an optimistic outlook about myself.	0.752

Developed by the author.

presents the survey’s variables, scales, outer loadings, Cronbach’s alpha, composite reliability (CR), including rho_a and rho_c, and average variance extracted (AVE).

Table 1 demonstrates the outer loadings, allowing the author to conduct factor analysis. It is confirmed that the OL values are 0.7 or above [79]. Because the JI1 and SE1 scales were below 0.7 and ineligible for additional investigation, they were eliminated (see Table 1). Additionally, Table 2 illustrates the model’s reliability and validity. According to our data analysis, the reliability tests (AVE, Cronbach’s alpha, and composite reliability) all met acceptable values (see Table 2).

Validity and Reliability Test

The results of the validity test of this study indicate that items of all the variables have resulted in greater than the threshold level of 0.5 [81], which confirms the validity of our study (see Table 2). Thus, according to Heale & Twycross [82], validity is the extent to which a concept is accurately measured in this type of research. Moreover, ref. [83] pointed out that validity and reliability always go hand in hand in this type of study. Therefore, the results of the reliability test, including the items of all the variables in this study, show that the alpha scores (see

Table 3. Reliability statistics.

Variables	Cronbach’s Alpha	Cronbach’s Alpha Based on Standard Items	Number of Items
CEO burnout	0.839	0.883	6
Job insecurity	0.816	0.878	4
AI awareness	0.839	0.891	5
Self-esteem	0.830	0.882	5

Developed by the author.

) are above the threshold level of 0.7 [84]. Based on these results, we confirm that our study has consistency over time.

3.3. Profile of Responders

The demographic breakdown of the respondents amounts to 615, and there is a slight majority of males (55.8%) over females (44.2%). Stratification of age reveals that the greatest number of respondents belongs to the 41–45 age bracket (34.8%), followed by the 46 and above (25.4%) and the 36–40-year-olds (20.2%), with the under-25s making up the smallest percentage (0.3%). According to the job category, most of them are in SME top management positions (64.2%), with SME CEOs making up 35.8%. As for work experience, almost half of the participants have between 7 and 9 years of experience (49.6%), followed by 10 years and above (36.1%), while a mere 1.0% have between 1 and 3 years of experience. Educational level is diverse, with the largest percentage holding a Master’s degree (35.8%), followed by Bachelor’s degree holders (31.7%), PhD holders (22.3%), and high school diploma holders (10.2%). This is a highly qualified, predominantly experienced, and senior professional sample. Complete information is included in the table (see

Table 4. Item, characteristic, frequency, and proportion (%).

Item	Characteristic	Frequency	Proportion (%)
Gender	Male	343	55.8
	Female	272	44.2
Age	Under 25	23	3.74
	25–30	37	6.02
	31–35	82	13.33
	36–40	124	20.16
	41–45	214	34.8
	46 and above	156	25.37
Job Category	CEO	220	35.8
	Top manager	395	64.2
Work Experience	1–3 years	61	9.9
	4–6 years	82	13.33
	7–9 years	305	49.6
	10 years and above	222	36.1
Education Level	PhD degree	137	22.3
	Master’s degree	220	35.8
	Bachelor’s degree	195	31.7
	High school diploma	63	10.2

Developed by the author.

).

4. Analysis and Results

The study’s inner and outer models were assessed using PLS-SEM (SmartPLS version 4). The following criteria were employed to assess the outer model: model fit, discriminant validity, AVE, Cronbach’s alpha, CR, and OL. R-squared, Q-squared, and hypothesis testing make up the evaluation of the inner model.

4.1. Discriminant Validity

We have conducted discriminant validity to check for correlations between our study measure of the construct and measures of other, unrelated constructs. With regard to this, we implemented the heterotrait–monotrait (HTMT) test for the discriminant validity of the data. A common threshold for discriminant validity suggested by Henseler and colleagues [85] is equal to or less than 0.85. Values above this threshold are not considered as discriminant validity. A low correlation coefficient (near 0) shows high discriminant validity, which means the constructs are distinct. This is typically assessed by checking correlation matrices, cross-loadings in a factor analysis, or the heterotrait–monotrait ratio (HTMT) in a structural equation modeling (SEM) analysis.

4.2. Structural Model Assessment

While evaluating the hypotheses of the study and inner model,

Table 5. Discriminant validity.

Variables	A	CR	AVE	AI	JB	JI	SE
AI	0.830	0.891	0.674	0.821	* 0.577	* 0.503	* 0.506
JB	0.803	0.864	0.561	0.492	0.749	* 0.699	* 0.605
JI	0.844	0.883	0.645	0.412	0.593	0.803	* 0.565
SE	0.820	0.876	0.657	0.430	0.516	0.477	0.811

Developed by the author. Note: AI = Artificial Intelligence Awareness, JB = Job Burnout, JI = Job Insecurity, SE = Self Estee, CR = composite reliability, AVE = average variance extracted. Diagonal values in bold are the square root of AVE. * Heterotrait–monotrait ratio (HTMT) is in italics.

puts emphasis on the direct, mediator, and moderator interactions. Robustness refers to the ability of the method to maintain good performance and provide reliable results despite violations of model assumptions [86]. Hence, the robustness statistics results indicated an acceptable level of a 95% confidence interval of 1.2. A key indicator of a robust test is its ability to keep the actual Type I error rate (the probability of incorrectly rejecting the null hypothesis) close to the nominal (e.g., 5%) level [87].

Hypothesis 1 (H1) investigated the direct effect of AI awareness on job burnout (β = 0.298, T value = 8.066, p < 0.000), indicating a highly significant effect. Hypothesis 2 (H2), which examines the mediation effect between AI awareness and job burnout, indicates a significant effect. Based on the statistical findings (

Table 6. Hypothesis evaluation.

H	Path	Β	T-Statistics	p-Values	Remark
H1	AI -> JB	0.298	8.066	0.000	Significant
H2	AI -> JI -> JB	0.194	6.846	0.000	Significant
H3	SE × AI -> JI	−0.206	6.191	0.000	Significant

Note: AI = Artificial Intelligence awareness, JB = job burnout, JI = job insecurity, SE = self-esteem.

), job insecurity mediates the relationship positively (β = 0.194, t = 6.846, p < 0.001). Finally, Hypothesis 3 (H3) demonstrates that self-esteem moderates the relationship between AIA and job insecurity negatively (β = −0.206), with a significant t-statistic of 6.191 and a p-value of 0.000. Therefore, all of the tested hypotheses are significant overall, denoting that self-esteem moderates the association between AI awareness and job insecurity, while AI awareness, directly and indirectly, affects job burnout through job insecurity.

Table 6 also displays the SRMR value of 0.067 that was discovered during the model fit test. A good fit index is defined as having an SRMR threshold value of less than 0.08, as Henseler et al. (2015) [88] confirmed.

4.3. Moderating Effect

Figure 2 shows that the link between executives’ AI awareness and their feelings of job insecurity depends on their level of self-esteem. When self-esteem is low (the red line), greater AI awareness is associated with a sharply rising sense of insecurity; at average self-esteem (blue line), the increase in insecurity is more modest; and at high self-esteem (green line), the relationship actually flips and is slightly negative, meaning executives with strong self-worth feel no rise—or even a small drop—in insecurity as their AI awareness grows. In other words, confidence in one’s abilities blunts, and can even reverse, the insecurity that usually comes from recognizing AI’s disruptive potential.

5. Discussion

The findings of this research generally support KM theory by revealing how employees’ knowledge of AI interacts with job insecurity and well-being. In particular, the findings indicate that AI experience causes psychological strain and knowledge behaviors (knowledge hiding and strategic use), which in turn affect knowledge-related organizational performance. In bringing these findings into relief with the extant literature [15,18,22,48,89], this paper identifies both similarities (e.g., for the evidence of knowledge workers experiencing greater burnout when the AI is more salient) as well as differences, including those relating to moderators pertaining to UAE organizational practices. These findings bridge KM theoretical constructs and I–O psychological mechanisms, lending support to claims that effective Knowledge Management in the digital context should not just consider structural or technological systems but also employee perceptions, distress, and personal resources.

Additionally, the findings of this study demonstrate a strong association between job burnout and AI awareness, validating the effect of AI-driven changes on worker well-being [22]. Additionally, AI awareness significantly contributes to job insecurity, aligning with [17,59]. They see changes behind AI as threatening their job security, inducing more anxiety and uncertainty about their jobs. Job insecurity plays a significant role in job burnout, which is consistent with previous studies [17,21,90]. Employees who are unsure whether they will be forced out of their jobs are at higher risk of emotional fatigue and decreased motivation [91]. Job insecurity has also been identified in the research as a mediator between awareness of AI and job burnout [17,51]. It shows that with increased AI awareness, the perception of threats to security in the labor market increases, which acts as an antecedent of burnout [54,90,92]. The findings also indicate that self-esteem is a factor that affects the relationship between awareness of AI and job insecurity. Overall, the employees, who are more aware of AI and have higher self-esteem, show lower job insecurity [31,72,74].

5.1. Theoretical Implications

This study offers several theoretical contributions to the literature on AI awareness, job insecurity, and occupational well-being within the framework of Knowledge Management (KM) and Industrial–Organizational (I–O) Psychology.

First, it reveals that digital transformation has basically altered the dynamics of the workforce across numerous industries, generating both opportunities and challenges for individuals in the workplace [22,93]. Awareness of this technology’s potential impacts triggers psychological responses that highly affect employee well-being. While AI awareness produces plausible positive implications, such as new employment opportunities and increased income for certain skilled positions, it simultaneously increases substantial concerns regarding technological disruption and job displacement [1]. These concerns, mainly relevant to the excessive implementation of AI systems and encompassing psychological stressors, are referred to as “technostress” [94]. It is a kind of stress that people experience when interacting with and gaining awareness of technology [95]. Prior findings have revealed that digital transformation might alter work systems in organizations. Nonetheless, such a work environment would offer ambiguous factors to the career patterns of the employees. Personnel who have high AIA are frightened of losing their jobs and becoming jobless shortly. Thus, they will be less enthusiastic when dealing with their present duties, making them feel uninterested, exhausted, and burned out due to the effects of digital transformation [18,22].

Secondly, the study demonstrates that job insecurity has become a major psychological stressor in the contemporary workforce [22], and it is a particular assessment of “a re-marked risk to the steadiness and constancy of engagement as it is currently practiced” [48,96]. According to knowledge-based theory, employees who handle their organizational performance via the effective use of their knowledge are labeled as knowledge workers [15]. However, the extensive use of AI knowledge might lead to employees’ distress, triggering a behavior of knowledge hiding to maintain their job security. Additionally, effective leadership fosters cooperation among staff to achieve the goals of the corporation and reduce knowledge hiding [48,97]. In other words, concerns about the job insecurity of employees tend to increase as they learn more about how AI usage might transform workplaces. In particular, AI awareness encompasses the belief that digital transformation poses a threat to one’s professional advancement; higher awareness is associated with greater adverse behavioral reactions [1]. The emergence of job insecurity, which is an immediate result of burnout, can show up in various forms. Decreased professional organizational goal achievement, exhaustion, and detachment trigger as a consequence [89]. This relationship has been empirically supported in research that demonstrates how employees’ perceptions of job insecurity, which are affected by increasing degrees of AI awareness, lead to job burnout [18,89]. Burnout caused by job insecurity leads to a lower level of performance and commitment as well as increased absenteeism in the workplace [98]. Additionally, greater awareness of AI’s application has been strongly linked to adverse well-being outcomes of workers, such as job burnout, depression, and decreased satisfaction with work, particularly in service sector contexts [18,98]. AI may impact individuals’ professional development, and they may fear that it will take the place of their present positions. Moreover, AI tools are more effective than humans, which might decrease the sense of personal achievement of employees and promote burnout in their current jobs [22].

Job insecurity, often resulting from burnout, can manifest as reduced goal achievement, emotional exhaustion, and detachment [89]. Empirical evidence indicates that employees’ perceptions of job insecurity, exacerbated by AI awareness, increase burnout [18,89]. Burnout further reduces performance and commitment while increasing absenteeism [98]. Greater AI awareness has also been associated with negative well-being outcomes, including depression and lower job satisfaction, particularly in service sector contexts [1,18,98].

Third, the moderating role of self-esteem as a personal resource, which is a significant addition to the original Job Demands–Resources (JD-R) model, incorporates psychological characteristics like persistence and having the ability to effectively handle the external environment [99,100]. Personal resources that are a person’s positive self-perception of resilience and their capacity to influence and manage their immediate environment can be employed as a moderator in the JD–R [101]. Self-esteem in particular acts as a buffer in the JD-R model, reducing the correlations between negative outcomes and stressors at work. Individuals who possess higher self-esteem typically perform better than those with lower self-esteem. Studies show that a lack of self-esteem can significantly lead to psychological issues like stress, depression, and thoughts of suicide [101,102]. In the workplace, having high self-esteem can help to lessen the stress from work demands, generating better overall experiences [101,102]. In addition to providing support, self-esteem can help employees make the most of the resources they have at work, which can improve their well-being and engagement [38,101].

Moreover, the UAE is enthusiastically taking advantage of AI to fulfill its ambitions and achieve the goals of establishing a fully technological governance. The UAE has made AI a top priority to make sure that the country develops as a developed and advanced nation [103]. A lot of rewards have been allocated to create and maintain happiness in workplaces. For example, the Happiness @ Work Award was released in the UAE in 2018 in cooperation with Forbes Middle East. This reward is recognized to identify corporate objects in the country that provide their employees the highest degrees of comfort and happiness [104]. According to a study, several factors in the UAE specify the well-being of employees, such as work overload and long working hours [105], job characteristics [106], leadership style [107], autonomy [108], and resource availability [109]. Organizations should implement integrated approaches that protect employees’ mental health by decreasing risk factors in organizations, promoting positive aspects of work and employee strengths, and addressing existing mental health issues regardless of the reason [110], which is in line with Sustainable Development Goal 3.

Finally, this study contributes to stakeholder theory by demonstrating how internal perceptions of managers (AI awareness and job insecurity) influence external legitimacy efforts, such as ESG compliance. Steadfast with stakeholder theory [36,42,111], leaders’ perceptions of technological disruption influence how they respond to stakeholder expectations and sustain organizational trust [112]. Executives perceiving AI as a threat may resist transparent reporting, potentially undermining ESG quality. Stakeholder-driven legitimacy pressures shape firms’ digital transformation strategies, providing a useful theoretical lens for understanding organizational responses to sustainability uncertainty [43,113].

5.2. Managerial and Policy Implications

This study yields several actionable insights for executives, managers, and human resource (HR) practitioners seeking to ensure that digital transformation remains both effective and psychologically sustainable.

First, leaders must recognize that the widespread adoption of AI and smart technologies is inevitable and transformative. Digital transformation challenges five core management domains: organizational culture, leadership, talent management, technology integration, and decision-making [114]. Organizations should foster an adaptive and innovative culture that promotes psychological safety, continuous learning, and openness to change. Building digital literacy and AI competence across the workforce will reduce uncertainty and empower employees to view AI as a supportive tool rather than a threat.

Second, strategic organizations must develop strategic HRM systems that emphasize sustainable talent development. This includes identifying skills gaps, providing reskilling programs, and creating digital career pathways that encourage growth across organizational boundaries, aligning with the concept of boundaryless careers [115]. Embedding AI literacy and adaptive competencies into training can enhance resilience and confidence, thus reducing burnout and job insecurity [18,22,116].

Third, managers should proactively monitor and manage employees’ perceptions of job insecurity. Transparent communication about AI implementation plans, fair performance evaluations, and participatory decision-making processes can alleviate uncertainty. Furthermore, integrating psychosocial support systems—such as coaching, employee assistance programs, and mindfulness training—can enhance emotional well-being and align workplace practices with SDG 3.

Finally, leadership practices should emphasize human-centered AI adoption. By aligning AI initiatives with ethical principles, inclusivity, and employee well-being, organizations can ensure that digital transformation supports sustainable human capital rather than depleting it. This perspective bridges technological innovation with I–O psychological sustainability, promoting long-term organizational health. Digitalization can benefit employees if managed alongside supportive structures and organizational resources [3,5].

5.3. Limitations of the Study and Future Recommendations

First, the sample of the study data was only collected from SME CEOs and executives in the UAE, a region that is a unique topographical area with a spectacular, distinctive culture in comparison with other Western cultures [18]. This might restrict the generalizability of the study’s results. Future research may include data from different countries or cultures to validate the results. Second, the study just measured one mediator. There might be other variables that buffer the relationship between AIA and employee burnout, such as CSR (Customer Social Responsibility), self-efficacy, and technostress. Third, the study findings may not be well-organized due to a very limited sample size, which consists just of executives, managers, and SME CEOs in the UAE context. Future research should consider a broader sample, such as organization employees, to gain various points of view. Lastly, this study offers actionable pathways for a sustainable digital transformation aligned with SDG 3 (Good Health and Well-Being); however, scholars might examine how AI awareness is related to psychosocial dynamics by concentrating on additional SDGs, such as SDG 9 (Industry, Innovation and Infrastruc-ture), which may frame studies regarding how AI awareness among employees leads to innovation initiatives within the industry. Moreover, future research should accentuate SDG 4 (Quality Education) by exploring how the KM system, while integrating with AI awareness, may enhance employees’ learning and digital literacy.

6. Conclusions

This study discloses how AI awareness affects job burnout among SME CEOs and senior managers in UAE firms. Drawing on data from 615 participants, we examined a complex model in which AI awareness intensifies burnout while job insecurity mediates the relationship. Moreover, self-esteem moderates the association between AIA and job insecurity.

Theoretically, this study enriches the Knowledge Management (KM) and Industrial–Organizational Psychology literature by demonstrating that knowledge acquisition—particularly when centered on emerging technologies such as AI—can paradoxically deplete managerial psychological resources. Awareness of AI’s disruptive potential triggers cognitive and emotional strain as leaders attempt to make sense of their evolving professional roles. This process underscores the necessity of organizational mechanisms that facilitate knowledge assimilation, sensemaking, and psychological adaptation during technological transitions. By integrating self-esteem into the model, the study also extends the Job Demands–Resources (JD-R) framework, highlighting personal resilience as a critical resource in mitigating the stress associated with digital transformation. Practically, the results invite executives and HR strategists to integrate mental health protections and clear plans with respect to skill transition into AI roadmaps while embedding self-esteem building within their knowledge management platforms. By concentrating on the mental health implications of digital change, the study is in line with SDG 3 (Good Health and Well-Being) and highlights that the long-term sustainability of digital transformation safeguards the mental health of the managers and executives, fostering higher engagement, productivity, and trust across the firm.