First, in terms of advancing our understanding of the relationship between changes in the work environment and leader outcomes, we call for more research on both top-down and bottom-up work design.

Top-down changes include aspects such as a company-wide introduction of return-to-office policies (e.g., Amazon, cf. Jassy, 2024) or, conversely, the adoption of virtual-first strategies (e.g., Dropbox, cf. Bersin, 2024). Such organizational shifts pose a dual challenge for leaders: First, leaders — unless they are at the very top — often bear the responsibility of communicating these changes and managing subordinates’ potential dissatisfaction. They must “make them work” in practice, which includes reconciling organizational mandates with employee expectations. Second, the work design itself creates new demands for leaders. For instance, hybrid work environments have introduced novel coordination challenges, requiring leaders to balance fairness, performance, and engagement across in-person, remote, and hybrid teams. These shifts heighten the importance of seamless communication (Bartsch et al., 2021) and necessitate additional technological competencies (Tigre et al., 2023). To address this two-fold challenge, future studies could investigate how hybrid or remote work policies influence both the demands placed on leaders to communicate and implement these changes — often facing resistance or dissatisfaction from employees (van der Velde & Gerpott, 2023) — and the specific challenges leaders encounter as they navigate and lead effectively within these evolving structures, particularly in comparison to how such policies affect the work demands of employees (cf. Lamovšek et al., 2025).

Beyond top-down changes, the dynamics of bottom-up work design also merit further exploration (Parker et al., 2025). Remote and hybrid work arrangements have amplified bottom-up work design dynamics by increasing employees’ autonomy to proactively shape when, where, and how they do their jobs (Wessels et al., 2019). Yet, the frequency and type of employees’ job crafting behaviors could inadvertently reshape the leader’s own work design. For example, employees who frequently craft their tasks to maximize flexibility might create ripple effects, reducing the leader’s autonomy by requiring additional coordination efforts or adjusting their own task focus. Conversely, employees who craft their roles in ways that reduce their dependency on the leader might afford leaders greater autonomy or free them to concentrate on more strategic responsibilities. Moreover, a pivotal yet underexplored issue relates to how leaders navigate the balance between fostering autonomy through employees’ bottom-up work design (e.g., job crafting) and ensuring alignment with their organizational goals remains a pivotal yet underexplored issue. Similarly, understanding the strategies leaders employ to provide diverse i-deals to team members while ensuring fairness and fostering a shared team identity could provide valuable insights into the increasing complexity of leadership roles.

Second, we notice that in discussions about leadership in remote work settings, the design of physical and digital characteristics of leaders’ work environment has so far been underexplored in its importance for leaders’ ability to influence others effectively.

Regarding the physical aspect, this oversight is surprising, given the substantial body of research demonstrating the impact of physical cues on leadership perceptions, particularly in shaping charismatic signals (for an overview, see Reh et al., 2017). Rooted in the concept of leader embodiment, this research highlights how physical cues, whether stemming from the leader’s bodily cues or their immediate surroundings, can enhance or undermine others’ perceptions of leadership effectiveness. Adding nuance to the debate, recent research additionally showed that physical environments that evoke awe in followers (e.g., grand lecture halls, stunning natural landscapes, or office spaces with expansive views) can amplify charisma attributions particularly for leaders who are not already known to be charismatic (McGuire et al., 2025).

While these insights have traditionally focused on physical co-presence, initial studies have begun to extend these ideas into virtual contexts but much remains to be explored. Offering a starting point, recent work on environment-based social class signals investigated how physical environments, as conveyed through videoconferencing, influence perceptions of a leader’s status and competence (Loignon et al., 2024). This research found that virtual backgrounds — ranging from authentic home settings to carefully curated or even fake ones — affect judgments of leadership competence and authority. While participants’ perceptions were influenced even when they knew the backgrounds were artificial, other scholars (Hardin et al., 2025) found that employees were more motivated to invest in their relationship with a coworker when they perceived the disclosure of the coworker’s private environment in a digital work setting as unintentional rather than intentional. This raises an interesting question: Could similar dynamics apply to leaders, and how might these two mechanisms together shape leader effectiveness? Curated status cues might increase employees’ perceptions of their leader’s competence and authority, while unintentional glimpses into private settings may enhance perceived authenticity and trust, potentially jointly influencing key leadership outcomes. Accordingly, future research should explore whether leaders benefit from such disclosures in ways similar to coworkers, or if these glimpses must be complemented by competence signals to meet the unique expectations of leadership roles (cf. implicit leadership theories, Eden & Leviatan, 1975; Schyns et al., 2011). Based on the findings, scholars could then derive valuable practical implications such as how leaders can strategically design their remote work setups to be perceived favorably and maximize social influence.

Such research could be complemented by studies that focus on the broader implications of digital characteristics of remote work environments for leadership tasks. To begin with, a critical area for investigation is identifying design features of digital work environments that help leaders maintain team cohesion and identification — two challenges that have consistently been identified as central in increasingly virtual work settings (Krug et al., 2021; Maurer et al., 2022). For example, in virtual meetings, team members may be more likely than in face-to-face settings to engage in backchannel communication — a form of multi-communication where additional interaction streams run parallel to the primary discussion. This can include using instant messaging services like WhatsApp during a Zoom meeting to exchange messages with selected colleagues, which can have different purposes (e.g., supporting each other and fostering a sense of connection or task-focused reasons; Mu et al., 2024; Stephens, 2012). As research (Sarkar et al., 2021) with Microsoft employees during the COVID-19 pandemic showed, on the one hand, such parallel chats can facilitate coordination around shared resources and enhance a sense of belonging for those who are a part of them. However, they can also be distracting, overwhelming, and create information asymmetries, potentially excluding those not involved in these side conversations. Understanding how to effectively manage these parallel interactions could offer valuable insights into how leaders can foster inclusive digital communication environments while ensuring that informal exchanges enhance rather than hinder team cohesion and collaboration.

Third, we also see opportunities to update or extend existing leadership theories by considering alternative mechanisms and leader-facilitated shared experiences that may gain relevance in digital or hybrid work design contexts. To illustrate what we mean, consider the realm of gig work, where scholars have critically discussed and then affirmed that the construct of “organizational citizenship behavior” persists despite the absence of direct colleagues and ongoing employment relationships (Moorman et al., 2024). Yet, the conceptual mechanisms underpinning this behavior in gig work contexts likely differ from those in more traditional work environments (Moorman et al., 2024). Similarly, structural changes in a leader’s work design may influence leader effectiveness by amplifying the importance of task-related and social dynamics.

Research found that changes in work characteristics toward more virtuality can influence task-related shifts in team coordination processes (Bartsch et al., 2021; Maurer et al., 2022). As teams increasingly operate outside of co-located work, traditional coordination mechanisms (e.g., spontaneous in-person discussions or informal check-ins) are replaced by more structured, technology-mediated processes such as real-time collaboration tools to maintain workflow synchronization (Maurer et al., 2022). In that regard, Bartsch et al. (2021) reported that virtuality amplifies the need for explicit task structuring to ensure smooth coordination. On the other hand, these structural changes may also create space for shared task-oriented leadership. That is, if leaders actively encourage team members to take on leadership in coordinating task execution, this can positively influence employees’ perceived productivity and satisfaction with their leader (Mayer et al., 2023), who in turn may share even more leadership with team members (Klasmeier et al., 2025).

In terms of social dynamics, research on remote work highlights perceived isolation as a central challenge (Gajendran et al., 2024), which may increase leaders’ job demands if they take a more active role in counteracting employees’ loneliness. In this regard, it would be interesting to study interventions such as leader-facilitated shared experiences (e.g., virtual vs. face-to-face team-building activities) that could influence relational mechanisms such as strengthening trust and reducing loneliness, which are essential for bridging employees’ remote work characteristics with performance outcomes (Golden et al., 2023; Wang et al., 2021). At the same time, leaders themselves may also be at risk of loneliness through two ways: first, leaders lose spontaneous peer support, making it harder to share challenges and seek advice; second, digital communication can make leadership feel more transactional, limiting the depth of relationships with employees (Lam et al., 2024). As such, research should explore how to mitigate leader loneliness in remote work settings, including interventions that foster peer connection among leaders, facilitate informal networking opportunities, or integrate relational support structures into leadership development programs. In sum, initial evidence indicates that both employee and leader loneliness may act as crucial mediator in the relationship between remote work design characteristics and leadership effectiveness. Yet, while these examples for mediators are largely empirically driven, we encourage scholars to also think broadly about overlooked theoretical perspectives to pursue research that examines a wider range of alternative mechanisms in hybrid or remote work contexts.

While not yet the case, the human prerogative over three key leadership functions — task-, relations-, and change-oriented roles — is likely to diminish (Van Quaquebeke & Gerpott, 2023). In the new workplace reality, AI will first steadily complement an increasing number of leadership functions. Since communication lies at the core of organizational coordination, one task that remains for human leaders will then be navigating the challenges introduced by AI copiloting (Van Quaquebeke & Gerpott, 2024), which particularly involves the augmentation of workplace interactions through generative AI tools that combine Large Language Models with conversational interfaces. While AI offers significant potential to enhance communication quality at work, it also risks undermining trust and respect if not managed responsibly. Imagine writing a well-crafted email sharing personal issues with your leader, only to receive an obviously AI-generated response in return — how would that make you feel? Probably not too great; you might feel objectified and not seen (cf. Granulo et al., 2024). With these risks in mind, we suggest that the increasing adoption of AI copiloting means that leaders will soon play a crucial role in guiding employees to use it effectively, ensuring that human agency remains central.

To shape this emerging future, we call for both descriptive and explanatory research. First, we encourage leadership scholars to conduct field studies that explore how AI copiloting is already deployed in contemporary organizations and which leader experiences exist in using such systems. For example, just as Dell’Acqua et al. (2025) have shown how generative AI as a “cybernetic teammate” reshapes team collaboration by influencing expertise integration, emotional engagement, and performance, we see a need to explore the implications for those who lead such human-AI constellations. That is, which new challenges and opportunities arise for leaders who must coordinate and legitimize AI-supported work? For example, do leaders actively signal their involvement in AI-supported processes to ensure that employees perceive AI as a tool guided by human judgment rather than as an autonomous agent? Are they transparently communicating the rationale behind AI-generated decisions or embedding themselves in interactions that AI initiates? Assuming not all leaders are equally motivated or equipped to engage in this new orchestration role, future research could even ask how important the human leader's presence really is (i.e., would the outcome differ if the human leader had not been involved at all?).

Second, we also encourage experimental studies to better understand how AI copiloting can be designed and utilized in ways that help leaders and employees preserve the relational and respectful aspects of workplace communication. Such research could explore, for instance, the impact of customization tools in generative AI — such as pre-prompt settings that allow leaders to adjust the tone, formality, or complexity of AI-generated messages — on employees’ perceptions of their leaders. Moreover, experimental designs could manipulate the frequency and circumstances of human interactions with leaders to determine how much “real” contact is necessary to mitigate the AI objectification effect and sustain trust in leader-follower relationships (Granulo et al., 2024).

In the distant future, we anticipate that AI will take over core leadership functions (i.e., not only task-related guidance but also relation- and change-oriented leadership) as it becomes increasingly adept at addressing employees’ fundamental psychological needs for autonomy, competence, and relatedness (Van Quaquebeke & Gerpott, 2023). Broadly speaking, the ongoing debate among scholars and practitioners responds to this anticipated shift from two distinct perspectives: an optimistic view and a pessimistic one.

To put it bluntly, the pessimistic perspective foresees a descent into the “emptiness machine,” where AI-driven technological advancements fail to improve working conditions. Instead, the advent of AI drives higher expectations for productivity without necessarily enhancing employee well-being (e.g., Lindebaum & Langer, 2024). This aligns with the productivity paradox (Brynjolfsson et al., 2019), where technological improvements do not yield proportional gains in productivity, as well as rebound effects, where efficiency gains lead to increased usage or demands, ultimately negating the intended benefits (Ertel & Bonenberger, 2025). While this latter perspective may be relatively new to leadership scholars — who have predominantly understood rebound effects in terms of how leaders rebound in their careers after forced breaks or significant life events (e.g., Dotlich et al., 2011; Sonnenfeld, 2007) — we see this as suggesting promising avenues for future research. Specifically, scholars could explore how rebound effects manifest when AI copiloting amplifies certain leadership tasks, thereby potentially fostering over-reliance on AI (Buçinca et al., 2021; Klingbeil et al., 2024) or challenging leaders to prevent a decline in the average creativity of their teams (Doshi & Hauser, 2024). Another promising area pertains to exploring the unintended consequences of AI-driven efficiency on leader-employee dynamics. For example, does the support through AI raise performance expectations for leaders themselves and, in turn, for their employees (i.e., exacerbating collective experiences of burnout)? Or are there also positive aspects (e.g., leaders have more time to engage in strategic vision-related activities, thus allowing them to experience more sensemaking)? This leads us to the contrasting positive perspective, which offers a more hopeful outlook on the implications of AI increasingly taking over leadership functions.

The optimistic view suggests that AI’s integration into leadership could, from a pragmatic perspective, help address current challenges, such as the previously discussed shortage of suitable and willing individuals to assume leadership roles, as well as the reality that many human leaders fail to act as change-driving forces and instead exhibit poor leadership (Breevaart et al., 2024). Studies indeed repeatedly show that individuals react more positively toward AI as compared to human leadership — as long as they are not aware that it is an AI (Stock-Homburg, 2025; Yin et al., 2024). While it may seem tempting to withhold this information, doing so is not advisable, as trust takes time to build but can be lost instantly. However, similar to previous technological advancements, initial AI aversion will likely diminish as employees gradually become accustomed to interacting with AI (Van Quaquebeke & Gerpott, 2024).

Beyond these functional considerations, by taking over routine tasks, the integration of AI into leadership may indirectly prompt human leaders to rethink their roles. Traditionally, as leaders climb the career ladder, they tend to lean heavily toward operational execution rather than “real” leadership tasks that focus on employee engagement (DDI, 2016). Against this backdrop, the anticipated shift driven by AI could be interpreted as an opportunity to rediscover the original essence of leadership, such that AI, by automating routine and execution tasks, could free leaders to embrace this original understanding of leadership. We propose, as depicted on the left side of Figure 1, that this freedom challenges human leaders to concentrate on three areas (cf. Gerpott & Jansen, 2025): 1. Acting as decision-makers in situations of uncertainty (i.e., organizational level), 2. shaping employees’ identity through communal co-presence (i.e., interpersonal level), and 3. mastering the art of maintaining appropriate distance toward themselves and AI (i.e., intrapersonal level).

Each of these areas implies, in our view, the need for research on future leadership development (see the right-hand side of Figure 1). Assuming that everything capable of optimization will be optimized, we anticipate that leadership development programs will increasingly leverage AI to personalize learning experiences. While some express concerns that such a reliance on AI could lead to deskilling (Lindebaum & Ramirez, 2024), we contend that a critical differentiation of two qualitatively different forms of learning can bring greater nuance to the debate. Particularly, leadership development encompasses varying depths of learning, which can be broadly categorized as incremental learning and transformative learning. Incremental learning focuses on refining or adjusting specific leadership skills and techniques, enabling leaders to improve within established modes of work and thinking. In contrast, transformative learning represents a deeper process, requiring leaders to fundamentally redefine their sense of self and approach to leadership by critically reflecting on ingrained habits of mind through the view and emotional resonance of others (Kim et al., 2023). While AI has significant potential to enhance incremental learning, it is less suited for fostering transformative learning processes (Kerschreiter & Gerpott, 2024).