ARTIFICIAL INTELLIGENCE CHANGES MANAGERIAL WORK MORE THAN LEADER IDENTITY: A QUALITATIVE STUDY

ARTIFICIAL INTELLIGENCE CHANGES MANAGERIAL WORK MORE THAN LEADER IDENTITY: A QUALITATIVE STUDY Feruza Abdivalieva Saint Petersburg State University, Graduate School of Management Saint Petersburg, Russia Abstract. This study examines how artificial intelligence affects leader identity and the practical content of managerial work. Based on nine semi-structured interviews with leaders and managers involved in AI implementation, it analyzes whether AI produces a deep transformation of leaders' self-understanding or primarily changes how the leadership role is enacted. Theory-informed thematic analysis reveals a persistent distinction between the core of leader identity and changing role expectations. Participants continued to define leadership through responsibility, direction, support, communication, judgment, and employee development. At the same time, AI altered leaders' tools, speed, autonomy, and work intensity, while AI literacy became a new criterion of competent and legitimate leadership. The paper concludes that, at the current stage, AI transforms the conditions under which leadership is performed and evaluated faster than it transforms leaders' deeper self-understanding. Identity change is therefore incremental: practices and expectations shift first, with deeper selfdefinition potentially following later. Keywords: artificial intelligence; leader identity; role expectations; leadership; AI literacy; legitimacy; managerial work. Introduction Discussions of artificial intelligence and leadership frequently assume a radical transformation of the leader's role. AI can perform analytical, communicative, and coordinating functions, support decision-making, and generate managerial recommendations. These developments raise a question that goes beyond changes in tasks: does the use of AI alter how individuals understand themselves as leaders? Leader identity refers to the incorporation of a leadership role into an individual's selfconcept. It is constructed not only internally but also through interaction with others. Individuals claim a leadership role, while others grant or withhold that role through recognition and the authorization to lead (DeRue & Ashford, 2010). Technological change may therefore affect identity indirectly by reshaping expectations concerning a leader's competence, responsibility, and legitimacy. Conceptual research suggests that AI may gradually assume functions traditionally associated with leadership, including analysis, the generation of decision options, and personalized communication (Van Quaquebeke & Gerpott, 2023). Empirical studies, however, more often indicate an increasing complexity of managerial work than the replacement of the manager. Leaders need technological awareness, adaptability, and the ability to transform organizational processes (Hossain et al., 2025). In AI-integrated teams, managers face expanding contextual, task-related, competence-based, social, and relational demands (Koponen et al., 2025).

It is therefore important to distinguish between changes in work and changes in identity. Leaders may adopt new tools, perform more tasks, and meet new standards of competence while continuing to define leadership through familiar categories. This study examines how leaders describe the influence of AI on their self-understanding and how changes in leader identity differ from changes in managerial practices and role expectations. Methodology The study used a qualitative, exploratory, and interpretive design. The empirical material comprised nine semi-structured interviews, each lasting approximately one hour, with leaders and managers directly involved in AI implementation or use. Participants were based in Russia, Uzbekistan, Kazakhstan, and the United Arab Emirates and represented banking and fintech, engineering, energy, manufacturing, construction, marketing, IT recruitment, and digitalproduct development. The organizations differed in size and AI maturity, ranging from initial experiments to the regular use of AI in key organizational functions. Purposive sampling was supplemented with elements of snowball sampling. The interviews were conducted in Russian. Questions concerning leader identity addressed managerial self-definition, changing expectations of good leadership, the influence of AI on authority and expertise, and situations in which AI-supported decisions were or were not recognized as legitimate. The data were anonymized using codes R1-R9 and examined through theory-informed thematic coding and cross-case comparison. The analysis did not seek statistical generalization; instead, it identified recurring interpretations of change. All interview excerpts presented below were translated from Russian into English by the author. Findings The first finding concerns the stability of the substantive core of leader identity. Participants continued to describe leaders in terms of responsibility, direction, team support, coordination, communication, honesty, and final decision-making. R3 stated: "At a fundamental level, the manager's role [...] has not changed with the arrival of AI. What has changed is the toolkit and the approach." Even with intensive technology use, leadership was not understood as a competition between a person and AI. The system was perceived as an extension of an existing managerial function. R5 offered a similar interpretation, describing AI implementation as an "entirely organic and logical process" because a leader had already been responsible for helping the team work faster and more effectively. This response points not to an identity rupture but to the incorporation of a new tool into an established narrative of the self as a leader. When leadership was already understood as finding useful means to achieve a collective result, AI could be assimilated without revising the individual's basic self-understanding.

The second finding concerns changes in the practical enactment of the role. AI shortened individual operational steps, accelerated search and material preparation, and expanded leaders' access to analysis and prototyping. However, time saved did not necessarily reduce workload. In several cases, it was reallocated to additional tasks, thereby increasing work intensity. R4 associated AI use with greater autonomy and fewer approval cycles but emphasized: "The ability to use artificial intelligence [...] does not define me [...] either as a person or as a leader." Autonomy and efficiency could therefore increase without producing a deep identity change. The third finding is the continued centrality of human and relational functions. R2 explained: "For me, being a leader is about conversations - listening, hearing people, and talking to them. AI does not help me with that, but it frees up time." In this account, AI does not replace leadership interaction but may indirectly create resources for it. This distinction matters because technological efficiency may either displace communication through a higher volume of tasks or support it by reducing routine work. The outcome depends on how the leader reallocates the time saved. The fourth finding is a change in the criteria of competent leadership. Although the core of identity remained relatively stable, participants argued that leaders now need to understand AI's capabilities and limitations, formulate tasks for a system, assess its outputs, and explain the technology to employees. AI literacy was moving from an innovative advantage toward a professional norm. R8 expressed this expectation sharply: if a manager does not understand AI and its capabilities, "people may look at you as if you were a caveman." The statement captures a shift in expectations: a lack of technological awareness may call a leader's modernity and competence into question even when traditional leadership functions remain intact. The fifth finding concerns legitimacy. The interviews did not reveal a consistent perception that AI use by itself undermines managerial authority. On the contrary, a leader could appear more effective and open to development when the technology produced visible benefits for the team. At the same time, uncritical copying of AI outputs, failure to understand their content, and attempts to hide behind the authority of the system were viewed as potential threats. Legitimacy therefore depends not on whether AI is used but on whether the leader transforms a technological output into an independently justified judgment and accepts responsibility for it.

Taken together, the findings indicate incremental rather than radical identity change. AI first alters behavior and the conditions under which the role is enacted: decision-preparation practices, speed, breadth of analysis, autonomy, and knowledge requirements. It then changes the criteria by which others assess managerial competence. Deeper self-definition may remain stable. In other words, technology changes the ground on which leader identity is claimed and granted faster than it changes the internal answer to the question, "Who am I as a leader?" Discussion The findings clarify the distinction between leader identity and role expectations. The claiming-granting perspective explains identity as an alignment between an internal claim to leadership and recognition by others (DeRue & Ashford, 2010). AI does not necessarily disrupt the internal claim. Participants continued to regard themselves as responsible human agents who provide direction, support others, and make decisions. The external grounds for recognition, however, are expanding to include AI literacy, critical judgment, and the capacity to organize responsible human-technology interaction. The conclusion that change is gradual is consistent with studies showing that AI increases the demands placed on managerial work. AI integration raises cognitive complexity, expands verification work, and requires leaders to balance productivity, ethics, and human relationships (Koponen et al., 2025). At the senior-management level, openness to AI, strategic technology use, cross-level collaboration, and the management of ethical risks are becoming increasingly important (Bevilacqua et al., 2026). These changes are substantial, but they do not by themselves demonstrate the replacement of human leadership. The theoretical contribution of this study is a proposed sequence of change. Tools and role requirements change first; the conditions of legitimacy change next; only after these shifts may a deeper transformation of leader identity occur. This sequence avoids two extremes: technological determinism, which assumes that AI immediately creates a fundamentally new type of leader, and a simple continuity argument, which overlooks changing criteria of competence. For practice, leadership-development programs should combine technical AI literacy with the preservation of the human foundations of leadership. Managers need skills in formulating tasks for AI, verifying outputs, understanding risks, and designing human-AI interaction.

Communication, accountability, employee support, and contextual judgment remain equally important. AI competence becomes part of legitimate leadership not as a substitute for these qualities but as a new condition for expressing them effectively. Conclusion The study shows that, at the current stage, AI changes managerial work more rapidly than it changes leaders' deeper self-understanding. Participants described greater speed, autonomy, analytical capacity, and work intensity, as well as new expectations of technological competence. At the same time, they continued to define themselves as leaders through responsibility, direction, support, communication, and final human judgment. The central conclusion is that identity change is incremental. AI first restructures the practices and criteria through which leadership is enacted and recognized. The absence of a reported "change in who I am" therefore does not mean that no transformation is taking place. What changes are the conditions under which leaders must demonstrate competence and legitimacy. The study is limited by its small and heterogeneous sample, its reliance on leaders' selfreports, and its cross-sectional design. Future research could compare leaders' selfassessments with employees' perceptions and examine change over time, especially as organizations introduce autonomous AI agents and expand technological participation in managerial decision-making.

Adabiyotlar, References, Литературы: 1. Bevilacqua, S., Ferraris, A., Matzler, K., & Kuděj, M. (2026). Strategic leadership at high altitude: Investigating how AI affects the required skills of top managers. Journal of Business Research, 205, 115878. https://doi.org/10.1016/j.jbusres.2025.115878 2. DeRue, D. S., & Ashford, S. J. (2010). Who will lead and who will follow? A social process of leadership identity construction in organizations. Academy of Management Review, 35(4), 627-647. https://doi.org/10.5465/amr.35.4.zok627 3. Hossain, S., Fernando, M., & Akter, S. (2025). Digital leadership: Towards a dynamic managerial capability perspective of artificial intelligence-driven leader capabilities. Journal of 32(2), 189-208.

4. Koponen, J., Julkunen, S., Laajalahti, A., Turunen, M., & Spitzberg, B. (2025). Work characteristics needed by middle managers when leading AI-integrated service teams. Journal of Service Research, 28(1), 168-185. https://doi.org/10.1177/10946705231220462 5. Kragt, D., & Day, D. V. (2020). Predicting leadership competency development and promotion among high-potential executives: The role of leader identity. Frontiers in Psychology, 11, 1816. https://doi.org/10.3389/fpsyg.2020.01816 6. Peifer, Y., Jeske, T., & Hille, S. (2022). Artificial intelligence and its impact on leaders and leadership. Procedia Computer Science, 200, 1024-1030.

7. Van Quaquebeke, N., & Gerpott, F. H. (2023). The now, new, and next of digital leadership: How artificial intelligence (AI) will take over and change leadership as we know it. Journal of 30(3), 265-275.