Digital Transformation of the Accounting Profession at the Intersection of Artificial Intelligence and Ethics

1 Introduction

Alongside climate change, globalization, and demographic changes, the digitization of society and the economy is one of the driving trends of our time. Digital change is driven by technological progress and society’s ability to adapt to and integrate these changes. Artificial intelligence (AI) is a key component of this digital transformation that is revolutionizing various sectors, including accounting. As digitization advances, massive transformation processes are expected within the labor market, the education system, companies’ production processes, and consumer preferences. Globalized and competitive markets may help accelerate the innovation, adoption, and application of new technologies.

The potential benefits of AI in the accounting profession are increasingly being discussed in academia and the corporate world. The emphasis on innovation and intelligence has increased due to the COVID-19 pandemic and the implementation of sustainable development objectives. AI technologies may directly or indirectly alter economic activities, offering new opportunities for growth and efficiency. As markets change, organizations must adapt to stay competitive. The accounting profession is one industry that is undergoing significant changes to adapt to market requirements. While accounting tasks were traditionally performed exclusively by employees, automation powered by AI is now transforming the process by streamlining and optimizing repetitive activities (Bruner et al., 2020; Richins et al., 2017; Yoon et al., 2015). There is an ongoing global discussion on digitization and the performance of databases, data centers, and mobile networks. Notably, many decisions are increasingly based on the data and information processed within these systems. However, when machines or AI systems make decisions, the consequences for humans can be significant. For example, a mistake in financial reporting by an AI system could result in financial losses or legal problems for a company. Currently, AI systems are not yet capable of making comprehensive, autonomous, moral decisions, such as determining what is good or bad, or right or wrong. Given that the accounting profession is about making decisions and applying professional judgment, adapting technologies that act based on a predetermined model is not simple. Moral principles are input by humans and implemented in algorithms, which can ultimately lead to morally grounded actions (Etzioni & Etzioni, 2017). Despite this automation, human oversight and access controls should remain available to support AI-powered systems. As the number of small- and medium-sized businesses using AI-based systems increases, companies must consider whether the use of AI allows for responsible and fair action to assess future risks, such as loss of confidence or restrictions due to regulations (Basri, 2020). Uniform norms and standards are being developed to make AI applications safe, ethically justifiable, and connectable for midsize companies. These norms and standards should consider fundamental ethical principles (Azman et al., 2021; Shi, 2020).

International competitiveness is often associated with economic policy discourse and generally desirable situations. However, if one devotes oneself to this term’s more precise conceptual design, then only very rough and abstractly formulated concepts are revealed in many cases. In a narrow economic sense, competitiveness is generally understood as the ability to expand or maintain prosperity. Broader interpretations can consider aspects such as sustainability, health, or other sociopolitical goals beyond the gross domestic product. For example, green competitiveness refers to the ability of a country or organization to compete in the global market while maintaining or improving environmental quality (Gu & Yan, 2017).

Research conducted in the accounting profession underscores the importance of maintaining human control and access options to ensure the correct application of technologies in the field. This emphasis on human oversight in the face of technological advancement is empowering, as it ensures that humans remain in control of the ethical application of AI in the accounting profession. Considering the facts mentioned earlier, ethical methods and terminology must provide adequate support for the achievement of objectives in this sensitive area. The gravity of this issue cannot be overstated. One discipline that can support the development of ethical methods and terminology is business informatics. Business informatics is essential to understanding and informing the ethical and moral aspects of AI implementation. Questions about what is ethically and morally necessary, and thus subject to a deliberative process, must be determined and operationalized in management systems (Zhao et al., 2023). AI-based processes are used more frequently for the routine work done by accounting professionals because AI can process these tasks faster and more accurately. In the context of the development of new technologies, education will also play an increasingly important role, as curricula are created to support new market requirements and to train future professionals according to these requirements (Appelbaum et al., 2017; Eachempati et al., 2021; Kovalenko et al., 2021; Loebbecke & Picot, 2015; Losbichler & Lehner, 2021; Qasim & Kharbat, 2020). Furthermore, artificial neural networks are inspired by the human brain’s ability to learn complicated patterns in data by changing the strengths of the synaptic connections between neurons (Hinton, 2018).

The primary objective of the study is as follows: to empirically investigate the impact of digital transformation – particularly the integration of AI technologies – on various dimensions of quality of work life among professionals in the accounting field.

Based on the theoretical model of Easton and Van Laar (2013) and supported by the previous research on work and technology (Korunka et al., 2008; Kundi et al., 2021; Zaman & Ansari, 2022), the study focuses on six key constructs: job satisfaction, general well-being, workplace stress, job control, work–life balance, and working conditions.

Recent empirical studies have highlighted how AI tools influence these dimensions in the technical, psychological, and organizational domains. Abu Afifa et al. (2024) provide evidence from Vietnam that AI adoption can improve employee engagement and satisfaction. Chen et al. (2024) show how digitalization affects well-being and autonomy among accounting employees. Abdallah et al. (2025) compare the implementation of AI in the private and public sectors, underscoring the role of organizational readiness and workforce perception in shaping outcomes. Almaqtari (2024) emphasizes that effective IT governance structures reduce resistance and facilitate smoother transitions. From a human capital development perspective, Anica-Popa et al. (2024) stress the importance of competency building to meet the evolving demands of AI-supported professional practice. Similarly, by applying the diffusion theory of innovation – which explains how, why, and at what rate new ideas and technology spread – Assidi et al. (2025) show that AI integration has a transformative effect on professional identity and employee control. Furthermore, Johri (2025) and Nguyen et al. (2025) explore how AI-enabled systems impact the accuracy of financial reporting and broader environmental, social, and governance performance, linking technical innovation with strategic decision-making. These insights provide a rich contextual backdrop for the empirical examination of how AI and digital transformation influence the quality of work life (QoWL) in accounting work environments.

This study aims to provide empirical evidence regarding how digitalization affects job satisfaction, general well-being, stress at work, homework interface, and job control and the proposed connection in the working environment/conditions. Our research investigates the influence of digitalization on a sample of Romanian employees in the accounting profession.

This article examines the path to operationalization and the link between innovative accounting techniques, ethics, well-being, and health literacy. The research aims to clarify the connection between these essential aspects and show that they must be harmonized to prove useful to the accounting profession and to reduce the risk that new innovative technologies, especially AI, may bring.

Against the backdrop of profound technological changes in the world of work – mainly due to the increasing use of AI in accounting – ethical and organizational issues are becoming increasingly relevant. Therefore, the challenges and objectives of this study require a solid theoretical foundation to systematically assess the complex interactions between digitalization, professional ethical requirements, and the quality of working life. The following section provides an overview of the relevant theoretical concepts and normative frameworks that serve as the basis for this empirical study.

While the introduction has outlined the broader motivations for investigating the role of AI and ethics in the accounting profession, a solid theoretical foundation is necessary to critically assess these developments. Section 2 presents relevant regulatory, technological, and ethical frameworks that inform and contextualize the empirical approach.

2 Theoretical Background

2.1 Regulations

AI technologies are on the brink of revolutionizing the accounting profession, promising significant economic and social benefits. These innovations, which have already demonstrated their potential in the environmental, health, public services, financial, mobility, home affairs, and agricultural sectors, are poised to significantly improve prediction accuracy, streamline operations, optimize resource allocation, and personalize services (The Artificial Intelligence Act, 2023).

The accounting field is governed by a multitude of standards and regulations that pertain to the ethics and deontology of the accounting profession, issued by various professional bodies at international, regional, and national levels, such as IAS (Service, 2020), GAAP (Board, 2020), IFRS (Standards, 2020), and FASB (Board, 2020). However, there is a noticeable absence of standards and regulations that address the ethics of AI in the accounting profession. This underscores the pressing need to update and adapt standards and regulations to accommodate new innovative technologies. It is not sufficient to simply regulate AI ethically; it must be guided and supervised. Regulatory bodies such as the Public Company Accounting Oversight Board (PCAOB) and the International Auditing and Assurance Standards Board (IAASB) are taking steps to establish oversight programs that can anticipate and respond to the risks posed by using these emerging technologies (IAASB, 2018, 2022a; PCAOB, 2018). In response to the rapid adoption of technology in audit practice, the IAASB has formed a technology working group to collect feedback from various stakeholders (regulatory bodies, supervisory bodies, accounting firms, academics, and professional bodies, among others). Stakeholders have observed that “data is being used differently than in previous audits” and have identified “legal and regulatory challenges” (IAASB, 2018). Although stakeholders do not view current standards as “broken,” there is a consensus that practice guidance (IAASB, 2018) is necessary to reflect the digital age in which the profession now operates, and regulators are calling for a review of standards in a “mode that reflects current technology” (IAASB, 2018). However, the use of AI in accounting also presents potential risks, such as data privacy breaches, algorithmic bias, and job displacement. Figure 1 illustrates the different AI-based technologies.

Figure 1

Overview of AI technologies (Source: IAASB, 2022b).

The Association of Chartered Accountants (ACCA, 2017) notes that “historically, machines simply ran programmes developed by humans. They were ‘doers’ rather than ‘thinkers’. Now, with sophisticated machine learning tools based on pattern recognition, systems can engage in discretionary decision making (World Economic Forum, 2018).” The latest 2018 World Economic Forum report provides an update on the foreboding use of Big Data and technology and the role of data in society. The Fourth Industrial Revolution has brought unprecedented opportunities and new challenges. To take full advantage of new technologies, we need to focus on what makes us human: our ability to learn new skills, our empathy, and our ingenuity. Creativity, in particular, will play a significant role in shaping the future. By actively learning new skills, we can adapt to the changing technological landscape and ensure broad progress.

2.3 The Ethical Challenges of AI

Ethics, the study of good behavior and morals, is a core discipline of practical philosophy. It is closely related to politics, legal philosophy, and economics, is based on morality produced by society, and is generally binding. One is unimaginable without the other. As a result of this relationship, there are arguments for and against certain behaviors. The ethics of AI deals with the ethical challenges that become visible through AI. Due to the many issues that have arisen with the use of AI in recent years, AI ethics has emerged as a new area of research and application in the field of applied ethics (Diab & El Hajj, 2024; Fülöp et al., 2024a, b; Wieringa, 2020).

Midsize companies increasingly use AI systems to simplify and automate processes and create a planning basis for future decisions. Therefore, firms must embrace AI ethics from the outset of system design as failure to do so can prevent a product from reaching the market or damage reputation and trust, thereby reducing acceptance among customers and society. The potential risks of not embracing AI ethics are high, and it is crucial for companies to understand and mitigate these risks (Etzioni & Etzioni, 2017; Murikah et al., 2024). Technical innovations often create new spheres of action for which known ethical forms of behavior do not provide an answer. Therefore, they must be discussed, reflected on, and adapted to the new requirements.

Ethical guidelines must be developed and verified for practical feasibility. Apart from legal requirements, companies should clarify early on which purpose AI should be used for. As the results of the study “Ethics in AI” show, employees evaluate the use of AI in a similar way to consumers. Values and guidelines provide security and allow the development and use of AI applications without having to decide for each application individually or cause a moral conflict (Agbon, 2024; Fülöp et al., 2024a, b; Munoko et al., 2020). Three areas of the company are particularly in demand: (i) the leadership and management level to initiate a company-wide AI strategy and code of conduct; (ii) customer-facing teams such as HR, marketing, and customer service to ensure transparent and ethical use of AI while providing full disclosure to customers (if requested); (iii) IT, AI, and data teams to ensure quality requirements for data and AI systems and document implementation (Anica-Popa et al., 2024; Assidi et al., 2025). Ethical standards are not equally important for every AI application. A basic rule can be generally established: The greater the risks associated with using a system, the greater the importance of ethical principles in the development process.

Companies are key players in the implementation of ethical design decisions. However, ethically relevant aspects of technological development are not only relevant to developers and engineers. The strategic alignment and commercialization concepts behind technology development are essential elements of a successful overall ethical concept. Any decision support system, whether heuristics, simple reports, or advanced AI, is human made and relies on updating the basic situation (data) and learning (expanding alternative courses of action). On the basis of the criteria of the European Commission, we have defined seven areas for the development of AI systems (Figure 2) (Cohen et al., 2020; Larsson, 2020; Palladino, 2021; Smuha, 2019; Ulnicane, 2022; Van Roy et al., 2021).

Figure 2

Development of AI systems (Source: Made by the authors).

The primary goal of app-related codes of ethics, particularly the “ethics briefing” of platform learning systems, is to provide guidance on value and action options for corporate practice. The central question is whether theoretical values can be effectively implemented in practice. Many companies are already prioritizing ethical development and application processes for AI systems. Several factors influence this: corporate culture and ethics, the market demand for “good” products, and compliance with legal standards. Ethics responds to emerging moral dilemmas by creating new fields of inquiry, such as area ethics. A prime example of this evolution is the ethics of AI within the realm of applied ethics. The central issue that arises is the moral implications of embedding moral abilities within AI-based processes (Etzioni & Etzioni, 2017). Furthermore, we must recognize the role of social responsibility and sustainable responsibility in achieving sustainable development goals.

From this perspective, the objective of ethics briefings and AI codes of conduct is not just abstract theorization but the creation of actionable feedback loops that connect moral frameworks to real-world applications. As the global community strives to achieve the sustainable development goals of the United Nations, ethical AI will be instrumental in promoting social responsibility, transparency, and long-term sustainability (Mohamed Riyath & Inun Jariya, 2024; Nguyen et al., 2025). This practical application of ethical frameworks should inspire and motivate us all.

Theoretical discussions have highlighted key aspects, such as the regulatory framework, the potential and limitations of AI, and the ethical challenges inherent in the digital transformation of accounting practice. It has become evident that technological innovations cannot be viewed in isolation but are always intertwined with organizational, social, and ethical dimensions. This interconnectedness should increase awareness of the complexity of the digital age.

To empirically test the theoretically derived relationships and thoroughly analyze the impact of digital technologies on the quality of working life in accounting, Section 3 provides a detailed presentation of the research objectives, methodological approach, and the data used.

Having established the theoretical underpinnings of AI integration and its ethical implications in accounting, we now turn to the empirical phase of our investigation. The following section outlines the research objectives, methodological design, and data collection procedures used to assess how accounting professionals experience digital transformation in practice.

3 Research Objective, Methodology, and Data

4 Results

A total of 113 questionnaires were included in the evaluation. The sample was based on exclusive accounting firms that use AI in their companies. The composition of the sample is presented in Table 1. Although the study provides valuable information on the impact of digital transformation on the quality of working life in the Romanian accounting industry, some methodological limitations must be noted. First, the sample size of 113 validated questionnaires was relatively small. Although this number is understandable, given the early phase of digitalization in Romania and the specific target group (companies with at least one AI innovation), the generalizability of the results is limited. In particular, only limited conclusions can be drawn from this sample regarding the industry or other national contexts. However, this study paves the way for future research in this area, offering a hopeful and interesting perspective for further exploration.

Today, new technological developments, particularly AI, are profoundly impacting our lives, both positively and negatively. This underscores the need for a balanced and comprehensive discussion that not only focuses on the advantages but also considers the potential risks and dangers (Figure 3). With AI’s global availability and transformative potential, such a balanced approach is not just desirable but essential.

Figure 3

Identified risk and digitalization problems (Source: Own research).

The use of AI is making great strides. Outstanding results have been achieved in various fields of application. The discussion of concrete application examples shows possible uses and opens up the perspective of possible extensions of financial processes. Two application areas show the most progress: accounting anomaly detection and financial data forecasting. In anomaly detection, visible financial transactions are automatically identified and evaluated based on transaction attributes, while automated financial data prediction supports transactional activities in corporate planning. Both process models achieve high added value for control after a short period of time. From automatic document recognition through account reconciliation to managing open items to automatic document pre-booking, an AI can take over redundant accounting tasks. However, certain activities require the knowledge of real experts, which an AI cannot keep up with. Automatic document entry and document processing support a smooth accounting process and eliminate time-consuming and sometimes unpleasant work from employees. Nevertheless, voices are raised pointing to the adverse effects of AI: above all, the loss of jobs. AI is a pioneering technology that is sweeping more and more areas of life. This leads to fundamental changes in the economy and society. In addition to technical, economic, and legal issues, this upheaval also involves moral challenges that are the subject of AI ethics. AI, with its potential to solve big problems with new ideas, inspires hope and optimism. With advances in AI and robotics, machines will increasingly make fundamental moral decisions that will affect our lives in the future. Machine ethics is a new discipline at the interface between computer science and philosophy that is concerned with developing ethics for machines instead of developing ethics for humans in dealing with machines.

Health is one of the most important things in life. In the context of the last year and a half, with the onset of the coronavirus, health has become even more at the center of society’s attention and, therefore, also of companies. Since the start of the COVID-19 pandemic, the number of hours employees worked remotely doubled. It should be noted that the mobile work is not possible in many sectors. In this context, the physical and mental health of employees is important. Organizations must also consider the social health of individual employees, that is, the state of social well-being at work. We inevitably find that digitization is both a curse and a blessing for our society. New responsibilities are emerging regarding the work environment and hybrid work models. Employers are subject to a digital duty of care, which, in addition to exemplary management behavior, includes a work environment that allows for personal responsibility. Furthermore, companies should be aware that perfectly functioning hardware and software are even more important than before the COVID-19 pandemic.

Similarly, employees should be aware of their responsibility and learn to use technology for their well-being. Fear of job loss through technology (Spencer, 2018; Vu & Lim, 2021) means that employees see their own jobs threatened as technological change advances and, among other things, allows greater automation. Furthermore, the extent of emotional exhaustion due to increased digitalization is influenced by the degree of use of information and communication technology (ICT) during leisure time at work. The growing link between digitalization and burnout is further reinforced by the growing use of ICT for leisure work. However, the low use of ICT reduces the risk of burnout, especially if there is a high degree of perceived digitization. This result suggests that self-restraint in the use of digital media for leisure time work purposes plays an important role in successfully managing digitization.

Our research was carried out using a rigorous approach to test validity and reliability and ensure the quality of the measurement instrument used. The validity of the construct was tested in two comprehensive steps. First, a convergent validity test was conducted, with the average variance extracted (AVE) calculated for each dimension of the QoWL. All AVE values were found to be above the accepted threshold of 0.50 (Hair et al., 2014), indicating sufficient convergent validity. Discriminant validity was tested by comparing the square root of AVE with the correlations between the constructs (Fornell–Larcker criterion). The results showed that in all cases, the square root of the AVE exceeded the intercorrelations, indicating good discriminatory power between the constructs. To test reliability, both internal consistency (Cronbach’s alpha) and composite reliability (CR) were calculated. The values were above the recommended threshold of 0.70 for all scales examined. Cronbach’s alpha was 0.889, and CR was 0.899, indicating high internal consistency of the scales. These results confirm that the survey instrument used is valid and reliable, providing a robust basis for empirical analysis. On the basis of our results on the model, we found that the factor loading value from the dimensions to the indicators is >0.5, as shown in Figure 4.

Figure 4

Output of the model results (Source: Own research).

The convergence validity test indicates a value of 0.513 for the QoWL, and a value of 0.5 is considered acceptable (Field, 2018; Hair et al., 2014), which is within the validation parameters. Based on the values of the discriminant validity test, the root results of the AVE in each dimension were higher than the root extracted from the average variance, so the discriminant validity criteria were met. The AVE value and the root AVE result for each dimension of the model can be found in Tables 2 and 3. The reliability of the construct had a value of 0.899 and Cronbach’s alpha value of 0.889, which is considered sufficient. On the basis of the results obtained after the application of the construct validity and reliability analysis, we can conclude that the dimensions of which the QoWL indicator is composed have been validated and are considered reliable. Thus, all of these dimensions contribute to the QoWL.

The results of the structural equation analysis provide differentiated evidence on the validity of these assumptions:

H1: Digitalization is a catalyst for job satisfaction: The data unequivocally demonstrate significant positive correlations between digitally supported work processes and perceived job satisfaction. These results are consistent with previous studies suggesting a boost in motivation through technological relief (Dhamija et al., 2019; Easton & Van Laar, 2013). Recent studies further reinforce this connection. Abu Afifa et al. (2024) and Zhang et al. (2025) illustrate how the targeted use of AI in accounting can enhance satisfaction, primarily by automating repetitive tasks and ensuring superior data reliability. Similarly, Awwad et al. (2024) accentuate the role of AI in enhancing the quality of work processes and their perception by employees.

H2: Digitalization is beneficial for general well-being: Well-being is positively correlated with the use of digital tools. The focus on temporal flexibility as a resource is particularly noteworthy. This confirms the potential of digitalization to contribute to psychosocial stabilization under certain conditions (Kundi et al., 2021; Palumbo, 2022). Chen et al. (2024) demonstrate that the digitized work processes can positively impact the emotional well-being of accounting professionals. J. Nair et al. (2024) also underlined that AI-supported processes can bolster the subjective sense of security and confidence in one’s skills, thereby enhancing the quality of life.

H3: Digitalization leads to increased stress levels in the workplace: This hypothesis was confirmed. The data showed that technological complexity, constant accessibility, and control mechanisms can lead to increased stress, a phenomenon described in the literature as “technostress,” which refers to the negative psychological and physiological impacts of technology use in the workplace (La Torre et al., 2020; Spector, 1988). Diab and El Hajj (2024) emphasize that ethical ambivalence and uncertainty in the use of AI represent additional stressors. Murikah et al. (2024) also show that a lack of transparency and algorithmic bias can trigger mistrust, which is emotionally stressful.

H4: Digitalization affects work–life balance: The analysis confirms that mobile technologies and home office models blur the boundaries between work and private life. This is perceived as a stress factor that reduces quality of life (Daniel, 2019; Williams et al., 2020). In line with this, Wang and Zou (2024) show that AI-supported performance tracking increases the expectation of constant availability. Rabbani (2024) also highlights an increase in psychological stress when digital technologies are implemented without appropriate regulatory or protective mechanisms.

H5: Digitalization is changing the experience of work control: The results reveal ambivalent effects. While some respondents reported greater self-control, others perceived the algorithmic specification of processes as a restriction. These divergent effects point to the relevance of individual and organizational moderators (Karasek, 1979; Kundi et al., 2021). Recent studies, such as those by Johri (2025), show that employees in highly digitized systems can experience a certain alienation when their autonomy is restricted. Conversely, Abdallah et al. (2025) and Almaqtari (2024) emphasize that targeted IT governance and participatory implementation strategies can significantly strengthen perceived control.

H6: Digitalization significantly influences perceived working conditions: The hypothesis has been empirically confirmed. Highly digitalized work environments are described as professional, modern, and resource-rich, which has a positive impact on perceptions of working conditions and safety (Zaman & Ansari, 2022). In this context, Pantea et al. (2024) provide empirical evidence that digital technologies in Romanian accounting practice significantly improve operational quality and the working environment. Similarly, Nguyen et al. (2025) and Al-Okaily (2024) show that technological infrastructure, combined with ethical regulation, is perceived as a stabilizing factor.

While the statistical results offer robust validation of the constructs measured in the study and confirm significant relationships between digitalization and various dimensions of QoWL, their practical meaning warrants further interpretation. In the context of the accounting profession, these findings suggest that the adoption of AI-enhanced tools improves operational efficiency and job satisfaction, primarily by reducing repetitive manual tasks and enabling faster, more reliable data processing. However, the concurrent rise in workplace stress and job insecurity signals the need for cautious implementation. These dynamics point to a shifting professional identity – where accountants are no longer merely data processors but strategic analysts and ethical overseers of AI outputs. Hence, beyond statistical confirmation, the results reflect a profession in transition, calling for targeted upskilling, ethical training, and participatory digital change management to support accountants in navigating this evolving landscape. The positive correlation between AI integration and improved working conditions also indicates that, when well-managed, digitalization can reinforce the perception of a modern, well-equipped, and professionally enriching work environment.

The findings shed light on both the benefits and challenges of digital transformation in accounting. To interpret these results meaningfully, the discussion that follows will integrate them with the existing literature, critically examine their implications, and reflect on their broader significance for theory, practice, and policy.

5 Discussion

The results show that all six dimensions contribute significantly to the QoWL, which encompasses various aspects such as job satisfaction, well-being, and perceived working conditions. In some cases, these dimensions interact strongly with each other. This interdependent structure underscores the complexity of digitalization experiences in practice and highlights the need for integrated management of digital transformation processes.

The European AI strategy, a pivotal initiative, aims to foster an AI ecosystem for innovation. It seeks to enhance the competitiveness of European research and industry, promote the diverse applications of AI in all social domains, and uphold standard European rules and values. The strategy’s focus on AI’s benefits for people, the common good, the environment, and the climate is paramount, offering a promising future for workplace dynamics and AI integration.

Quality of life at work is closely related to the environmental conditions in which we work and spend most of our time. Through a suitable work environment, we can achieve job satisfaction, which leads to a better work experience. The results obtained for job satisfaction indicate a high degree of job satisfaction that positively affects the QoWL (Dhamija et al., 2019; Singla et al., 2021). Job satisfaction is a relatively stable variable that expresses a person’s experience-based attitude toward their work situation. Job satisfaction affects many aspects of work life. For example, it is negatively correlated with turnover, meaning that dissatisfied employees are more likely to quit. Alternatively, satisfied employees will have far less absenteeism and are less likely to have workplace accidents. Thus, a link with job performance is also suspected.

Even though Herzberg et al.’s (1959) two-factor theory has little empirical support, it represents a plausible content theory of motivation with a good relationship with job satisfaction, the so-called hygiene factors, or dissatisfaction factors, which appear neutral when satisfied but produce job dissatisfaction when dissatisfied, and motivators or satisfaction factors, which have no effect when dissatisfied but lead to job satisfaction when satisfied. Of course, job satisfaction is determined by multiple factors. There are countless reasons why job satisfaction can increase or decrease: It can be affected by salary, the right to have a say, the behavior of the supervisor, the work schedule, the aspiration level of the employee, and many other reasons. Our results are consistent with the results found in the relevant literature, in which a high degree of satisfaction leads to a higher degree of QoWL (Dhamija et al., 2019; Easton & Van Laar, 2013; Fatehi et al., 2015; Zaman & Ansari, 2022). Stress and work are inextricably linked. Stress is a stimulating challenge, but too often it is a stressor. Based on this thinking, it is important to consider how stress influences job satisfaction. While employees would like to be challenged at work, too much stress or pressure at work can have a negative impact on satisfaction.

Stress at work can affect a person in a very individual way. In general, stress can be defined as the body’s reaction to an imbalance. This is due to external factors. Work stress affects both physical health and mental health. Those who are constantly stressed are quickly irritated and overwhelmed. The feeling arises that the requirements cannot be met. Stress at work has a negative effect on employees’ health and significantly affects our ability to concentrate and, therefore, our performance. Stress causes a hectic pace, leading to wrong decisions or a lack of focus. Our results confirm these statements about the negative effect on the QoWL, and these results are in line with those found in the literature (Contabeis et al., 2021; Spector, 1988; Surienty et al., 2014; Zaman & Ansari, 2022). This underscores the need for a stress-free work environment to ensure job satisfaction and QoWL.

The use of AI in the workplace can have widespread implications for workers’ safety and well-being. Mental well-being is not just a personal matter; it is integral to overall well-being and is crucial for our performance and competitiveness. By prioritizing mental well-being, companies can create a work environment where employees feel valued, balanced, and efficient, leading to increased productivity and success. It is not just a matter of employee satisfaction but a strategic imperative for businesses.

The pure physical stresses of the industrialization period have now given way to many other stresses that result, among other things, from the mixing of work and private life and from the use of new technologies. In addition, more people are overworking with little time for themselves. Such states of overload can lead to stress, an increased risk of illness, a lack of motivation, a loss of loyalty to the employer, internal resignation, and ultimately a change of job. All these factors have far-reaching – including financial – consequences for employers. However, employees who feel comfortable can improve their work performance and the competitiveness of their company just as significantly as the company’s success and competitiveness.

Mental well-being can be easily and surprisingly accurately assessed using a short questionnaire developed by the World Health Organization (WHO), a test with only five questions referred to as the WHO 5. As a rule, respondents only need about 2 min to answer the questions, which are all positively worded. Specifically, it asks to what extent the following feelings or moods have been prevalent in the past 2 weeks: good spirit, relaxation, activity and vitality, energy, and interest in things. Components of well-being (such as education, training, and work) and job satisfaction have a major impact on a person’s health. Conversely, work-related stress can promote illness. These connections have been well studied with respect to the development of psychological symptoms, particularly burnout. Therefore, we observed a direct link between well-being and the quality of life at work (Easton & Van Laar, 2013; Kundi et al., 2021; Palumbo, 2022).

Since the COVID-19 pandemic, more people have worked from home than ever before. Working from home allows employees to balance their private lives with their work. However, it also presents problems. On the one hand, it is easier to work overtime; on the other hand, the constant availability of colleagues must be viewed critically. Furthermore, the boundaries between work and leisure can disappear, which can negatively impact private life and thus the QoWL (Daniel, 2019; Williams et al., 2020). Control in the workplace is characterized by the degree to which an employee perceives the pressure to which they are exposed according to the required level of experience. Digitization and innovative technologies also come with new requirements and experiences in different fields. Accountants are no longer just gifted but must also be good computer scientists to understand and use AI-based innovations. Therefore, they become more stressed, which impacts their well-being and work satisfaction and thus their quality of life at work (Kundi et al., 2021). The COVID-19 pandemic also created long-lasting challenges and behavioral changes related to health and safety at work. Specific working conditions can positively and negatively affect employees’ health, well-being, and safety. Unhealthy working conditions are denoted as work stressors and protective factors as work resources. Working conditions in the financial services sector are not the same for all employees. Different industry trends in accounting, as well as the different job content of employees, suggest that outcomes within industries differ significantly. The same applies to the type of work employees do. Thus, an important connection between the quality of work determined by working conditions and the working environment was noted (Zaman & Ansari, 2022). As expected, all six dimensions influence the quality of life at work, and more than that, we note the interdependence between them. We consider the study and the results obtained as an X-ray of the QoWL in the accounting profession, which is undergoing a significant transformation from pencil and eraser to colleagues who are robots and take over part of the accountant’s activity.

Tasks such as sending invoices or repetitive administrative activities are tedious and take up a lot of time, usually more than companies can afford. Is not this time more sensibly invested in day-to-day business or customer service? Especially in accounting, intelligent programs offer valuable support, and some representatives are already in the market. In accounting and controlling, many repetitive tasks are tedious for employees. In addition, their execution means a time investment that is hardly justifiable today.

In summary, investments in digitization have a similar long-term effect on industries’ competitiveness as conventional capital stock intensification. In the short term, greater digitization goes hand in hand with higher productivity and lower unit labor costs. These competitive advantages can be reflected in higher export market share. The studies found no adverse effects of digitization on long-term employment development but signs of positive impetus for actual wage development. These results suggest that digitization leads to higher value production without large-scale, long-term employment releases, similar to the past technological developments.

Although the discussion underscored the need for ethical frameworks for the application of AI, the increasing prevalence of algorithmic decision support systems in accounting requires a more comprehensive exploration of specific ethical dilemmas. These dilemmas come to the forefront when the automated systems prepare or influence decisions that directly impact economic or social justice, for example, in risk assessment, audit selection, or the processing of sensitive personal data (Issa et al., 2016; Lehner et al., 2022; Munoko et al., 2020).

Key ethical areas of tension in accounting include, among others:

Transparency vs complexity: The often opaque decision-making logic of deep learning algorithms, which is complex for users to comprehend, poses an accountability challenge and contradicts the necessity for transparent decision-making (Cohen et al., 2020; Etzioni & Etzioni, 2017).

Automation vs human responsibility: Delegating critical decisions to AI systems can lead to an erosion of individual responsibility (“responsibility gap”), particularly concerning liability issues for incorrect decisions (Larsson, 2020).

Increased efficiency vs job insecurity: Although AI efficiently performs repetitive tasks, there is a risk of job displacement or loss, which has implications for social fabric and professional identity (Spencer, 2018; Vu & Lim, 2021).

Bias and discrimination: AI systems can reproduce or reinforce existing biases in data, for example, in assessing credit default risks or automated compliance checks (Ulnicane, 2022).

A sound ethical assessment of these dilemmas requires the consideration of different normative perspectives, including deontological (duty-based), utilitarian (consequence-orientated), and virtue-ethical approaches. Furthermore, it is recommended to follow established guidelines, such as the Organisation for Economic Co-operation and Development (OECD) AI Principles or the EU AI Act (2023), which call for human-centered, transparent, and accountable AI systems.

For accounting, this means, in concrete terms, that when AI-based systems are introduced and applied, ethical assessment systems must be implemented that reflect both regulatory and industry standards. Integrating such principles into training, practice, and technology design represents a key challenge for the future of professional ethics in the digital age. While the positive effects of digitalization in accounting – such as increased efficiency and decision support – are widely acknowledged, the potential adverse side effects that have recently gained prominence in academic debates should not be overlooked. In particular, the increasing reliance on AI systems raises questions about subjectively perceived job security and the psychological well-being of employees.

Numerous studies show that technological change has ambivalent effects on employees. On the one hand, the use of intelligent systems opens opportunities to relieve routine tasks and develop skills; on the other hand, automation of cognitive activities can trigger fears of displacement, role insecurity, and loss of control (Brynjolfsson & McAfee, 2014; Spencer, 2018). These developments particularly affect accounting activities, which are significantly transformed by data-driven automation. Another critical issue concerns the growing algorithmic control of work processes, which can create psychological pressure due to increasing performance transparency and monitoring mechanisms. Studies refer to this as “digital stress” or “technostress” (La Torre et al., 2020), which can be accompanied by symptoms such as exhaustion, excessive demands, and reduced job satisfaction. At the same time, there is a risk that a one-sided focus on efficiency and cost savings will undermine confidence in one’s own professional future in the long term. Uncertainty about which skills will be in demand can lead to fears about the future and a loss of motivation among employees (Tursunbayeva et al., 2021).

Therefore, a more balanced view of digital transformation processes requires greater attention to sociopsychological impacts, in addition to technological and economic potential. Future design approaches should focus specifically on transparent communication, participatory technology adoption, and the promotion of digital resilience, which refers to the ability to adapt and thrive in a digital environment, reduce fears, and strengthen trust in AI-supported systems.

The results of this study illustrate how digital transformation, mainly through AI, influences the working reality of accounting and financial reporting. At the same time, significant correlations between digitalization, work-related well-being, and ethical responsibility are evident. For example, increased use of AI can lead to improved efficiency but also to job insecurity, which can affect the well-being of employees and the ethical responsibility of the organization.

Building on these findings, Section 6 summarizes the key conclusions and draws implications for research, practice, and regulation. Drawing from the discussion of key findings and their contextual implications, we now conclude the study by summarizing the main insights, reflecting on the limitations, and proposing directions for future research. This final section aims to synthesize the contributions made and identify paths forward in the evolving landscape of digital accounting.

6 Conclusion

In the era of technological development, there have been rapid changes in professions and skills. There is the question of optimal operational management of the economy, high-tech industries, robotics, and the training of highly qualified competent personnel. In the next 10–15 years, there will not be one all-encompassing AI solution that can cover multiple user skills. Today, AI solutions are generally implemented for a single skill/function. Skills can be divided into three areas of application: language, imagery, and logical reasoning. Often, solutions consist of a combination of the three applications. As our analysis shows, there is no universally accepted definition of AI, Big Data, and related terms. Furthermore, the application of AI is likely to grow in the near future, although at different rates, depending on industry, region, and company size. Current and future use is expected to negatively and positively affect fundamental rights. However, we can ensure a secure and ethical future for AI with a comprehensive legal framework, political actions by the EU and its member states, and nongovernmental initiatives on AI’s ethical and legal implications.

The business world is changing. AI technologies and digital transformation in accounting are a reality that requires commitment. However, the question is not whether this is good or bad but whether companies are agile enough to take advantage of new opportunities. This includes freeing employees from repetitive tasks so that they can focus on the potential to create value together. AI will soon accompany many processes in medium-sized companies. However, employees need additional training to use AI and successfully develop safe management of AI-controlled processes. Machines and complex technological systems can functionally make and execute moral decisions. At the same time, new technological possibilities give rise to new moral problems that require ethical and moral reflections from the idea and conception phase. However, we must disregard such dark prophecies because the accounting profession is far from ending. AI should be considered the beginning of the profession’s renewal and will once again demonstrate its potential to adapt to recent changes in the business environment and changing management requirements. Accountants can benefit from intelligent systems because, using their capabilities, they will be able to solve three big problems: (i) supporting the decision-making process by providing better and cheaper data, (ii) providing deeper data analysis and new business insights, and (iii) focusing on more complex tasks, as employees can refocus their efforts while AI takes over more repetitive tasks. Thus, the involvement of accountants in the development of practical guidance and the effective governance of these technologies will be crucial. The new division of labor between humans and AI makes a company’s processes future-proof. Tax authorities are interested in accessing companies’ systems in real time and monitoring their accounting in this way. As a result, companies that still use manual processes are no longer competitive. This means that implementing AI in accounting is an absolute requirement.

Overall, the perceived digitization of one’s work environment is associated with health burdens for employees at the individual level. The harmful link to health is more substantial for emotional exhaustion. Moreover, a high degree of digitization is also associated with a more significant accumulation of work–family conflicts. Conversely, flexibility in the work situation is associated with a reduction in stress in the domains of individual health and family life. At the individual level, a good relationship with the manager is also associated with a reduction in ill health. Therefore, the results of our study confirm existing publications that associate digitization with adverse health effects and QoWL. As the digitization of the world of work will continue to advance in the next few years and permeate virtually all activities and sectors, successfully managing digitization at the individual and company levels is of particular importance.

The existing literature points to various influencing variables that may moderate the magnitude of health impairments caused by work-related stress and digital tasks. On the basis of these findings, our analysis of the successful management of digitization aims to identify influencing factors that can neutralize the potential adverse effects of digitization on well-being and health. However, as with any scientific work, the present study also has limitations. First, the measurements were based solely on the respondents’ perceptions, which means that the effects may be overestimated or underestimated. This applies to the perception of digitization and one’s health and well-being. Second, all study data were obtained through questionnaires and only at one point in the survey, meaning that the claims could only be correlative. Therefore, more studies that include a longitudinal design and that may resort to randomized interventions to allow causal claims should be pursued in the future.