Employees' adaptation to human–robot cooperation: a career adaptability perspective Open Access

Working with robots is becoming a reality across various industries, requiring significant adaptation from employees accustomed to “traditional” ways of working (i.e. without robots). Examples can already be found in healthcare (Turja et al., 2019), manufacturing (Kopp et al., 2021), spaceflight (Hambuchen et al., 2021) or education (Belpaeme et al., 2018), and it is predicted that more industries will follow in the near future (Kim, 2022). While promising greater efficiency and productivity, the introduction of robots into workplaces represents a career-relevant transition, reshaping employees' roles and responsibilities, creating new skill demands and potentially redirecting their career paths (Kellogg et al., 2020; Zhang et al., 2019). These changes make employees' career adaptability, a psychosocial capacity to manage and respond to evolving career demands (Savickas and Porfeli, 2012), essential, particularly in the era of collaborative technologies where humans and robots share tasks and spaces. Unlike traditional robots, collaborative robots are designed to adjust their behavior to support or complement human labor. Yet, the interaction between humans and robots can take multiple forms. Kopp et al. (2021) categorize human–robot interaction (HRI) into four types: cell, coexistence, cooperation and collaboration, depending on the degree of interdependence, physical contact and robot characteristics (Kopp et al., 2021). This qualitative study focuses on employee adaptation to cooperative HRI in industrial environments, where humans and robots perform interdependent but distinct tasks within the same workspace.

Employee adaptation to working with robots refers to the process through which employees learn, adjust and sustain their behaviors to align with the new demands posed by the implementation of robots in their workplace (Liu et al., 2023). Successful adaptation occurs when a fit between the required behaviors and employees' capabilities to execute them is achieved. Labor process scholars have long emphasized how new technologies impact control, processes of deskilling or reskilling and worker autonomy, often producing tensions between efficiency gains and employee agency (Kellogg et al., 2020; Spencer, 2017). Further, the introduction of robots typically leads to skill disruptions; some skills become obsolete, while new ones are required in order to operate, monitor or collaborate with robotic systems (Zirar et al., 2023). Such disruptions require employees to be able to unlearn old routines and practices and acquire new competencies, making learning a central component of adaptation to working with robots (Ra et al., 2019).

In this context, successful employee adaptation to working with robots requires more than technical integration; it depends on how individuals engage and respond to change. Career adaptability theory offers a useful lens for understanding this process by shedding light on the individual-level adaptability resources (i.e. concern, control, curiosity and confidence) that employees draw upon to anticipate, engage with and grow from career-relevant disruptions (Johnston, 2018; Savickas and Porfeli, 2012). Similarly, workplace innovation, socio-technical systems (STS) and classic change management models (e.g. Lewin's three-stage model; Lewin, 1947) emphasize the importance of building employee readiness and aligning technological change with organizational practices and employee needs to facilitate employee adaptation (Pasmore et al., 2019). These perspectives converge on adaptation as a key condition for effective organizational change, emphasizing the need to prioritize and support employee adaptation throughout the process of robot implementation.

Among the many ways adaptability resources are expressed, workplace learning stands out as a critical adaptive response through which employees acquire the skills needed to adapt to working with robots (Jansen et al., 2025). Workers can engage in workplace learning in two complementary ways. First, they can engage in formal learning activities, which are structured and take place within educational settings specifically designed for learning (Tannenbaum et al., 2010). These activities can take place both on and off the job and include courses, training and seminars. Second, they can engage in informal learning, which is learner-initiated, takes place outside formal educational settings and encompasses learning from oneself (e.g. experimenting with new ways of performing work), learning from others (e.g. interacting with peers) and learning from non-interpersonal resources (e.g. reading professional publications). Informal learning is by far the most prevalent form of work-related learning (Cerasoli et al., 2018). Paradoxically, empirical research on the relationship between automation and employee participation in work-related learning consistently finds that individuals most vulnerable to job automation exhibit the lowest engagement in adult education and workplace training (Ioannidou and Parma, 2022; Koster and Brunori, 2021). From a career adaptability perspective, one explanation for this finding is that employees who view automation as a threat may feel less confident or perceive lower control (i.e. have fewer adaptability resources), making the prospects of adapting more difficult. As a result, they are less likely to invest in training or skill development, which represents a critical adaptive response (Jansen et al., 2025). This is particularly concerning, as failure to adapt to new skill demands may undermine long-term employability and hinder employees' ability to navigate career transitions, making supporting career adaptability essential not only for successful short-term employee adaptation but also for fostering sustainable careers in increasingly automated work environments (Chen et al., 2019; Chin et al., 2019).

Despite the growing interest in learning in the context of digital transformation (Charalambous et al., 2015; Lambrechts et al., 2021; Jacob et al., 2023), research on how to support employees in their adaptation to working with robots is limited, with some authors calling for more research on the topic (Kim, 2022; Zirar et al., 2023). The present study responds to this call by addressing the following research question: How do employees adapt to working with robots? By doing so, this exploratory qualitative study offers several contributions. First, it advances the literature on employee adaptation to robot implementation by offering highly needed empirical evidence on employees' firsthand experiences of working with robots. Specifically, results shed light on how employees respond to initial challenges, knowledge gaps and changing responsibilities when adapting to working with robots.

Second, it contributes to the literature on workplace learning by exploring how employees develop the necessary knowledge and skills to work with robots, and what their unaddressed needs are. Third, this study contributes to career research by advancing our understanding of how employees mobilize adaptability resources in response to technological disruptions such as robot implementation. By foregrounding employees' learning strategies and adaptation challenges, the study illustrates how career adaptability can support individuals in responding to, managing and growing through such changes. In doing so, the paper responds to calls for more empirical research linking technological change with career development. Finally, our findings provide actionable insights on how human resource professionals can effectively support employees in their adaptation to working with robots.

Sustainable careers research positions the person as the central actor in managing career changes, with career adaptability and career competencies as key building blocks for long-term employability and well-being (De Vos et al., 2020). From this perspective, adapting to working with robots depends not only on employees' technical competence but also on the personal resources they can draw upon to cope with challenges and navigate career changes.

Career adaptability refers to the psychosocial resources individuals draw upon to cope with career-related transitions (Johnston, 2018). At the core of the career adaptability construct are four key adaptability resources: concern, control, curiosity, and confidence (Johnston, 2018; Savickas and Porfeli, 2012). Concern involves anticipating how robots may reshape work; control reflects taking responsibility for engaging with working with robots rather than resisting it; curiosity encourages exploration of new ways of working with robots; and confidence supports employees in believing they can acquire the skills needed to succeed in robot-enhanced work settings (Savickas and Porfeli, 2012). For successful adaptation to working with robots, employees must activate these adaptability resources and translate them into adaptive responses, that is, concrete behaviors through which individuals cope with and adjust to changing conditions (Johnston, 2018).

Complementing career adaptability, career competencies capture the more tangible knowledge, skills, and abilities individuals develop to steer their careers (De Vos et al., 2020). As robotic technologies transform work processes, continuous learning and upskilling become central not only for short-term adaptation but also for sustaining careers over time. Workplace learning, whether through formal training or informal knowledge exchange, stands out as a key adaptable response for employees to develop the required skills to adapt to working with robots. Thus, in the context of adaptation to working with robots, career adaptability should go hand and hand with the development of career competencies. In this way, career adaptability provides the motivational and psychosocial foundation, while career competencies capture the practical skills and knowledge needed to meet evolving work demands, together enabling employees to handle the challenges of robotization and to build the skills necessary for long-term career sustainability.

Building on the person-centered dimension of sustainable careers (De Vos et al., 2020) and career adaptability theory (Savickas and Porfeli, 2012), this study examines how employees adapt to working with robots, focusing on their lived initial experiences with robots, the changes in skill demands they encounter and the learning needs that support their adaptation.

When we talk about employees' initial experiences of working with robots, we refer to the first interactions employees have with robots in a work environment, including their initial reactions, challenges they face and strategies they use to overcome them. These initial experiences play a pivotal role in shaping their adaptation, as early interactions between humans and robots are often marked by uncertainty, novelty and emotional responses that set the tone for future acceptance, trust and skill acquisition (Hoffman et al., 2022; Liu et al., 2023). While adaptability research highlights that employees who approach change with a proactive mindset and adaptive coping strategies are more likely to succeed (O'Connell et al., 2008; Ployhart and Bliese, 2006), this disposition is often tested during first encounters with unfamiliar technology. In fact, research on acceptance of robots in warehouses and manufacturing settings has found that employees' initial reactions to robots may vary from enthusiasm to fear of job loss, fear of physical injury or increased stress (Jacob et al., 2023; Kopp et al., 2021).

During initial interactions with robots, employees may struggle to understand how the robot works, what is expected of them and how their role is changing, all of which can create friction and cognitive overload (Charalambous et al., 2015). Research on adaptation to robots in industrial environments suggests that positive initial experiences with robots have the potential to reduce negative biases and fears, to foster positive attitudes and to increase employees' intention to work with robots (Baumgartner et al., 2022), whereas negative initial experiences can have the opposite effect, resulting in a dislike of robots and hindering employee adaptation (Savela et al., 2021). Especially because most employees will not be familiar with the robot being implemented in their workplace, positive initial experiences are essential to invalidate false preconceptions, reduce stress and facilitate adaptation to working with robots. Therefore, the first sub-research question of this study is:

To adapt to working with robots, employees need to acquire new knowledge and skills. In industrial settings, robots often take over easy to moderately difficult tasks from employees, primarily because they lack the capacity to manage more complex problems (Kim, 2022) resulting in an increase in cognitive demands and more challenging tasks for employees (Berkers et al., 2022). A recent review suggested that robotization is positively related to skill variety (i.e. the extent to which employees need to apply a variety of skills and capabilities to perform their jobs) (Jonczyk et al., 2025). Based on a review of seventeen papers, the authors concluded that robotization expands the skill repertoire employees must draw upon in their daily work. The need for the acquisition of new skills has also been emphasized by prior research, showing that working with robots requires employees to interact with and maintain new technologies, often involving knowledge of programming, automation or mechanical systems (Michaelis et al., 2020), as well as learning how to communicate with robots (Kim, 2022), coordinate actions (Correia et al., 2022) and share tasks (Komenda et al., 2021).

Alongside these findings, institutional and policy initiatives such as the European Skills Agenda (European Commission, 2020), OECD reports (OECD, 2023) and the Future of Jobs report (World Economic Forum, 2023) underline the need for a future-ready workforce. These initiatives consistently highlight, among others, technology literacy, analytical thinking and resilience as essential skills for navigating technological transformation. Rather than narrowing employee work roles, the implementation of robots increases the complexity and multidimensionality of human work, reinforcing the importance of both technical competencies and broader cognitive, social and metacognitive skills (Ra et al., 2019). Despite the ample recognition of these changing skills demands due to the implementation of advanced technology like robotics in the workplace, empirical research on how robotization specifically alters required skill sets remains scarce. Most existing studies address skill shifts in the context of labor market transformations, without detailing how these changes manifest at the individual level in employees' day-to-day work (e.g. Autor, 2015; Frey and Osborne, 2017). Yet, identifying specific skills needs is crucial for informing the design of targeted training programs. This urgency is further underscored by findings from the Future of Jobs Survey (2025), where 63% of employers cited skills gaps as a major barrier to business transformation and 85% of employers reported plans to prioritize employee upskilling.

Therefore, the second sub-research question of this study is:

Together with the need to develop new knowledge and skills, research on the implementation of robotic systems has consistently stressed the importance of employee training and education as they adapt to working with robots (Jonczyk et al., 2025; Panagou et al., 2023). Training is an important enabler in industrial human–robot collaboration, increasing employees' confidence in their abilities to work with robots (Charalambous et al., 2015) and supporting employee adaptability (Ployhart and Bliese, 2006). However, despite the recognized importance of upskilling the workforce, only half of the current employees have access to adequate training opportunities (World Economic Forum, 2023). This training gap poses significant challenges, as inadequate employee preparation for working with robots may increase anxiety and resistance (Arslan et al., 2021). Further, human resource and career development professionals are still struggling to understand training and development needs associated with employees' adaptation to working with robots (Kim, 2022).

Beyond formal training programs, workplace learning research highlights that much of the knowledge employees acquire emerges by engaging in everyday work activities and interacting with others in the workplace (e.g. by doing) (Tynjälä, 2008). Learning is a social process deeply embedded in the daily practices, relationships and contexts in which work occurs. Theories of situated (Lave and Wenger, 1991) and experiential learning (Kolb, 1984) emphasize that employees often develop new capabilities through hands-on experimentation, trial-and-error and reflection on their own and others' experiences. These informal, experience-based learning processes are central to employees' ability to adapt to technologically transformed work environments (Beane, 2019; Ferreira et al., 2017), yet remain underexplored in current robotization research. Therefore, the third sub-research question of this study is:

Given the relatively small amount of research on the topic, we chose an exploratory, qualitative approach suitable to understand participants' feelings, opinions and experiences (Hsieh and Shannon, 2005). A semi-structured interview methodology was selected as it allows to cover specific topics that are closely related to the research questions while at the same time granting the interviewees freedom to go more into depth if needed and possible (Braun and Clarke, 2006). We interviewed 18 frontline employees who recently started working with robots in their daily operations. Data were analyzed using qualitative thematic analysis (Braun and Clarke, 2006).

We looked for companies with recent or ongoing implementation of robots. Three companies located in The Netherlands participated in the study. Two companies were in the pilot phase of robot implementation, and the third company had robots integrated into daily warehouse operations for two years already. Two of the participating companies were from the logistics sector, and one was from the energy production sector. An overview of the companies is presented in Table 1.

Interviewees' work experience with robots ranged from two months to one year (M = 6.7 months, SD = 3.0). We interviewed 18 operators, of which 10 were men and 8 were women (see Table 2).

There are several examples of employees working with robots in the logistics sector. The following details two use cases of employees working with robots within our sample. While specific names have been removed for anonymity, the goal is to provide more detailed information to create a frame of reference for later discussed themes that are presented in the results section.

Both logistics companies in our sample implemented robotic arms in their warehouses to perform pick-and-place tasks. This highly repetitive task involves employees spending hours picking items from one location and placing them in another. Maintaining concentration and accuracy over extended periods is challenging, leading to higher error rates. To address these challenges, the companies deployed robots equipped with sensors and vision systems to accurately lift objects from moving conveyor belts. The robots operate autonomously but incorporate human-aware safety measures (e.g. slow down when a human is in close proximity). Additionally, the system includes an error notification mechanism: when an operational issue occurs (e.g. item misplacement, mechanical fault or inability to identify the object), the robot sends an alert message to the human operator, who is then responsible for solving the issue. When the operation runs smoothly, humans and robots work side by side, executing sequential steps within a shared workflow without direct physical contact. Based on Kopp et al.'s (2021) categorization of HRI types, this setup is best described as cooperation (i.e. humans and robots perform interdependent but distinct tasks within the same workspace).

The energy company deployed a quadruped robotic system to perform automated inspection tasks. Equipped with thermal infrared cameras and gas detection sensors, this robot autonomously navigates through predefined inspection routes, producing images of what it sees. These images and sensor readings are transmitted to human operators, who use that data to detect any hot spots or thermal anomalies in the equipment. The inspection task is dangerous for human operators, and the implementation of robots allows for the reduction of risks and increased inspection time and frequency. Human operators are responsible for monitoring the robot's progress, responding to alerts and taking manual control when needed for complex or unexpected situations. At the time of the interviews, the company was still in the proof-of-concept phase, requiring operators to accompany the robot during inspections, a practice that is expected to change as the robot implementation becomes more established.

Therefore, human operators and robots perform sequential work steps without requiring physical contact to accomplish the work. This form of HRI can be labeled as cooperation (Kopp et al., 2021).

The first author conducted all interviews between October 2023 and March 2024. The interviews lasted approximately 45 min and were held online or face-to-face, depending on the company's preferences and availability. Interviews were conducted in English.

The interviews started with a brief introduction of the study by the researcher, and respondents consented to recording the interview. In addition, participants were assured that the interview was strictly confidential and that results would be presented anonymously. To put the interviewees at ease, they were first asked to describe a typical workday, tenure and prior work experience. The following questions were based on an interview guideline that was previously designed based on the research questions of the study to ensure consistency in the information obtained ^[1].

Data were analyzed using the six thematic analysis phases proposed by Braun and Clarke (2006). In the first phase, to get familiar with the data, the first author transcribed, read and re-read each interview while initial ideas were noted down. Nvivo Software was used to analyze the interviews. In the second phase, preliminary codes were systematically generated across the data set, and relevant features were identified. During the coding process, we used a semantic approach, meaning that we did not look for anything beyond what participants said and identified the themes within the explicit or surface meaning of the data. After the first coding round, 72 codes were identified. The initial coding was performed by the first author.

However, another two co-authors independently reviewed and cross-checked 33% of the coding (i.e. six interviews, which is higher than the 10–25% suggested in the literature; O'Connor and Joffe, 2020) as a means of improving the analysis and generating dialogue between the research team. Following best practices in qualitative research, discrepancies were discussed collaboratively during a meeting until consensus was reached (O’Connor and Joffe, 2020). Next, codes were collated into relevant themes, and a thematic map of the analysis was created. Five themes were identified in this phase. In the following step, initial themes were reviewed to ensure they were coherent (internal homogeneity) but distinct from each other (external heterogeneity). During this process, some codes were combined, and others were removed, leading to 16 final codes, combined into three themes.

The qualitative thematic analysis revealed three main themes, each of which will be discussed below. These themes reflect employees' experiences of adapting to working with robots, capturing the challenges they faced (Theme 1), the evolving skill demands (Theme 2) and the ways they engaged in learning to work with robots (Theme 3).

Adapting to working with robots was not without challenges. During initial interactions with robots, employees felt stressed “initially a bit of anxiety because of course you don't know the machine and you don't know what can happen.”(LB2), scared “First time was very scary when I was with the robot”(LA7) and lost due to the lack of knowledge, especially when unexpected circumstances occurred; “if you don't know how to fix issues, then you have to ask all the time, and maybe they don't know either. It's not easy, and it's a little bit stressful”(LA8). This lack of knowledge frequently slowed down their work pace and forced employees to learn new things to be able to work successfully with robots (see theme 2). For example, one interviewee said, “You sometimes get errors on the computer, and you have to know what these errors are and what you have to do with them”(E1). In fact, employees reported learning as a central challenge when adapting to working with robots: “The main challenge was in learning and operating, for me personally because I don't have too much knowledge about the working of the robot”(E4).

In addition to challenges in learning and adapting, employees experienced different technical issues and struggled to “follow the pace of working”(LB2) of the robot. Employees faced challenges in ensuring smooth operations due to the robot's capabilities and limitations. For example, one employee said: “There are like 10% items that the robot cannot pick, and we are currently having issues sending only the perfect products”(LA10). Other challenges mentioned include problems with cameras and connection: “we had a few issues with the camera itself and the connection”(E2). Because of these technical issues, the robot created more work instead of reducing it “Due to all the problems, we had more work with it than it took out of our hands. So that was quite a problem”(E4), which increased employee stress. Further, observing their colleagues struggle influenced the perception of the robot among employees who did not directly interact with it, as reflected in this quote: “People saw how much connection we lost. I had to get the robot again, restart it, and everything, and if that happens too often, people are like, yeah, no, it doesn't work, we don't need it”(E4).

Lastly, employees faced challenges related to workplace design, noting that their environments were not well-suited for robots. While this issue was raised across different use cases, the nature of the challenges varied. For example, regarding the quadruped robot employed in the energy sector, one employee noted: “The accessibility is the biggest challenge that we had because it isn't designed for a robot. So, we had many places it could not reach on its own because there are doors and stairs, and it can't go up”(E3). In contrast, employees working with robotic arms in the logistics sector did not encounter such physical barriers. Instead, their challenges centered on the integration of the robot into the broader operational workflow and understanding how it functioned within the entire process.

Although most employees reported experiencing challenges when working with robots, some found robots easy to use and experienced no problems: “I don't have any problem with the machine because it's very easy to use. It's not complicated”(LA3). These differences can be explained by the resources employees had at hand to respond to challenges, as elaborated in the following section.

Sub-theme 1.1: Employees' responses to challenges.

When employees encountered challenges during initial interactions with robots, they first evaluated whether they knew how to solve the issue. If they didn't, asking for help was mentioned as the most common strategy “asking questions not only to my colleagues but also my manager and trying to understand their answers and to be more in touch with the robot”(LA6). Additionally, employees emphasized the importance of reporting issues to the managers “when it's a big or serious issue, I need to contact my supervisors to help me with it”(LB2). Furthermore, participating in training helped them become more confident, learn to solve problems and reduce initial stress. The following quote nicely summarizes this: “the anxiety changed when I learned a bit more about robots, and I know there's nothing to be afraid of […] The training helped reduce initial anxiety and gain confidence”(LB2). Similarly, another person mentioned, “Educating yourself is the way to remove any fear. The more I know what to do if there is some issue, I feel less afraid.”(LA5). In addition, viewing problems as opportunities for learning and improving the system motivated employees to persevere through challenges. For example, one employee mentioned, “Every time there was a problem, we got one step further”(E6).

Although seeking help, reporting issues and getting training helped employees address the challenges experienced while initially working with robots, not everyone adopted these adaptive strategies when facing problems. Some respondents described how colleagues reacted more skeptically or resisted engaging with the robots: “when we had problems with the robot, some people would leave the robot and then be like-we have other work to do- So if it was not working, they would stop it.”(E6). Another acknowledged: “[when we have problems] We stop that part, and we focus the rest of the activity on the other parts”(LA4).

How employees responded to the challenges they faced often depended on their adaptability resources. Many employees expressed curiosity and excitement about how robots would fit into their workflow, how collaboration with robots would unfold, and how robots could assist them in their jobs. For instance, one employee shared: “When I first heard about it, I was very excited. I was very curious as well. I immediately saw some potential, and I had some scenarios in my head that I was immediately working on. So, I was very excited and curious mostly”(E2). Others also expressed confidence and control, emphasizing their belief that they could adapt to working with robots: “I just approached it as something new that I need to learn that I don't have experience with, but I know I can manage it. So I thought it was really cool” (LA5). Employees who were curious, confident, and experienced a sense of control, key adaptability resources, viewed robots more positively. They saw an opportunity to make their work easier, reduce physically demanding aspects, and perceived the implementation of robots as an opportunity to learn and develop new skills: “For people that started at the beginning, it was an opportunity to grow and learn something else”(LA8).

Such dispositions fostered adaptive responses when challenges arose, such as engaging in training or seeking help.

By contrast, some employees expressed doubts and concern about job security, fearing replacement by robots. One worker expressed the fear of being replaced, saying, “So it means in the future there is no work for us […] Because the robot will take all my work over and I don't have anything to do”(LA2). In such cases, the fear of losing one's job triggered resistance behavior toward robots: “Some people say-just throw it in the river-because it's a robot and they see it taking over their jobs”(E6). For these employees, facing repetitive challenges (e.g. technical issues) lowered their motivation and engagement, leading to maladaptive strategies such as putting the robot aside and focusing on other tasks instead.

Despite this, having the opportunity to experience working with the robot and witnessing its benefits reduced skepticism among those open to experimenting and learning new things: “I had many questions, but after these were answered, I saw real possibilities for the operations”(E1), and “First, I was very, very skeptical about this robot. And right now, I am like, well, I saw what it could do, I am like okay”(LA10). This indicates that while adaptability resources such as curiosity, confidence, control, and concern initially shaped employees' responses to challenges faced, they were not static. Instead, these resources were developed through experience, as employees became more familiar with robots.

During the interviews, employees expressed that the introduction of robots resulted in an increasing division of tasks as “people became more specialized for a specific part” (LA8). Further, robots “increase the complexity of the work processes” (E1), even for those who do not work with the robot directly. One employee explained: “The robot adds complexity and things to consider for everyone, even though they might not work with it themselves”(E3).

This added complexity also demands a deeper “understanding of the flow of the work processes in combination with the flow of the robots”(LA3). Employees agreed that robots changed their job roles, shifting them to a more supervisory position. One mentioned, “My task will change to more of an overseeing role, like checking if the robot does its work well.”(E2). When talking about these changes in roles, an intriguing debate emerged during some interviews: Who is then in control, the human or the robot? Employees firmly believed that “Robots should be an extension of the work of people, not the other way around”(E5). However, employees mentioned that working with robots frequently made them feel like robots themselves: “I don't like it because it feels like you are a robot too”(LA3). Instead of reducing monotony and allowing them to focus on more engaging tasks, employees perceived the new tasks as equally boring and repetitive: “The robots picking everything for you, it's also a really boring job because you're just pressing a button and you don't do anything”(LB1).

Alongside these changes in the job characteristics, adapting to working with robots requires what one interviewee described as “another set of skills than the people in this job need now”(E5). Employees emphasized the need to develop new technical skills. This includes learning about the robot itself; “we would need a broad range of skills from many of the operators, such as basic skills of how to handle the robot.” (E1). These skill requirements varied depending on the type of robot used. Employees from the energy production company, who worked with a mobile robot, highlighted the need to understand “how can we create new routes that the robot should follow up, how to ensure what are the best practices on measuring, from what angle do I need to measure or take a picture to get the best results”(E1) all this required a deep understanding of the IT platform. Employees in this company also mentioned the need to learn to interpret the computer-based language, with one noting: “it's all computer language mostly, and you have to learn to analyze the data that you're seeing”(E6). Employees across all companies emphasized the importance of acquiring basic maintenance skills. As one participant explained, it was necessary to understand “how to take care of the robot, like basic maintenance tasks” (LA5). In the case of the mobile robot, this included practical tasks such as “learn how to replace the battery when it runs out”(E2).

Lastly, employees highlighted the need for safety awareness to prevent accidents and ensure smooth operations.

In addition to technical competence, employees need to develop problem-solving skills to effectively navigate the challenges that come with working with robots. As one interviewee emphasized, “we need to be able to think in solutions”(E1), highlighting the importance of a solution-oriented mindset when technical issues arise. Another added that employees should know “how to fix issues without creating another issue”(LB1), suggesting that the ability to troubleshoot involves more than just applying a quick fix. It requires careful evaluation of consequences and interdependencies within the systems. This type of problem-solving goes beyond routine responses and often involves creative thinking; “We did have problems almost on a daily basis. Certain things not working. Connections not working. Sensor equipment not working. So people need to be able to solve problems or be creative enough to try something out” (E1).

Lastly, several interviewees stressed the need to have a certain mindset. One operator said, “It's another skill set, of course, but also another interest somebody needs to have. You have to be interested and curious about the robot.” (E5). This intrinsic motivation to explore and understand the robot's functioning was often seen as a distinguishing factor between employees who embraced the technology and those who remained hesitant. Moreover, the ability to adapt to working with robots was linked to employees' openness to digitalization. One interviewee mentioned that it is important employees have “openness to digitalization, openness to trying new topics, and openness to taking the time to try and learn”(E1). This mindset also included a strong willingness to learn through exploration, as another employee explained: “You have to be open to learn, to find out its [the robot] limitations and what you can do with it or what you cannot.” (E6).

Employees reported three primary methods for acquiring new knowledge and skills required to work with robots: formal training, through peers and by doing.

As a starting point, employees received safety training “The first thing you are going to learn is the safety rules”(LA4) and technical training, “They start up with a bit of a presentation about how the robot works, what can it do or what can't it do, what can we do when we have a problem […]they explained some movements how to handle it”(E4). After the initial introduction to the safety rules and the robot, participants were encouraged to try out things “we just got the basics and then they said, now just play with it. Just test it a bit. See where you can use it”(E5). During this process, the trainer's help and guidance were essential for understanding what to do and how to do it: “In the beginning, I learned by looking at what the trainer was doing. After that, she told me, okay, you can try, and I will stay close to you to check if everything is correct; if you have some information that you don't know, you can ask”(LA3).

Most interviewees reported that a big part of the learning occurred “by doing, just by doing”(E1) and by trying things out: “It is more like trial and error. It's about keep on doing it and get more experience, and then the job becomes easier.”(LA2). Some interviewees reported learning by doing as their preferred way of learning :“You have to learn by doing it, and that's how to get best to know it”(E6). If problems appeared, employees were encouraged to ask questions not only to the trainer but also to their colleagues “I don't always go to the trainer, but I also ask my colleagues”(LA2). In fact, the value of peer learning was highlighted by different employees who stated that knowledge about how to work with robots is acquired informally via their colleagues, “train a set of people who are then allowed to train other people, and that's how we share our knowledge about the robot.”(E2). However, although the importance of peer learning was highlighted by all companies in our sample, only one had a system in place to facilitate the transfer of knowledge among employees, “we always try to put shifts together. So experienced people are with less experienced people, and they can help out”(LA5). Within this company, training was organized at different levels, as explained in the following quote: “We have different levels of training. For example, a person who only has level one training for the machine cannot fix the machine.”(LA6). Another person added: “Level one, this is for people who can only press buttons and don't touch anything. Level two, the person can clean the machine or fix when you have some problem”(LA7). For effective peer learning, interviewees emphasized the importance of creating an environment where making mistakes is allowed and people are encouraged to ask for help.

Sub-theme 3.1: Gaps and opportunities in current practices

While employees highlighted the importance of training in adapting to working with robots, they also identified areas for improvement. Although the training provided a solid basis for their learning, employees felt unprepared when unexpected situations occurred. One employee commented, “The training missed unusual situations. We learned how to operate, how to do the job, but sometimes there are some unusual situations, and I don't know what to do”(LB2). Other employees expressed that training was frequently rushed and did not provide enough time for guided practice: “At the end of the training, you practice with the robot for 15 min in total. That's way too little […] the training should be longer or with more robots or fewer people so you get real guided hands-on time”(E4). Interviewees expressed that this made them feel lost when returning to work and believed that more practice time would help them feel more confident when working with the robot alone: “It would have helped to have more practice, to repeat everything a few more times just to make sure next time you are alone, no one can help you, and then you're able to do”(LB1). Further, some employees mentioned forgetting the knowledge gained during training after a few days. They suggested implementing periodic training sessions to reinforce what they learned and advance their knowledge and skills rather than relying on one-time training, which they found overwhelming. “It would be nice to have more extensive training but not in front because it's too much information at that time. So, it's good to have basic training and then move from there”(E6).

In addition to the prior feedback on training, employees felt communication about robot implementation was scarce, increasing their questions, concerns and feelings of uncertainty. Employees emphasized the importance of transparent communication to get everyone on board and ensure a smooth implementation. As one interviewee shared; “They just need to communicate what's the real purpose of the robot, what kind of jobs they see it taking on, and show to people that it's going to be an extension of their job and not a replacement”(E5).

Overall, employees were satisfied with the training provided by their companies and felt that it prepared them to do the job: “After the training, I felt that I could do it.”(LA2).

However, the interview results revealed that employees would benefit from additional training time, more training moments, knowledge on solving unexpected situations and more transparent communication from the beginning.

Drawing on career adaptability literature, this explorative qualitative study examines how employees adapt to working with robots. We interviewed 18 operators about their initial experiences while adapting to working with robots (i.e. challenges faced, and strategies used) and explored how employees acquire the required knowledge and skills and what their under-addressed needs are.

While previous literature provides theoretical overviews of key factors in workplace robot implementation (e.g. Charalambous et al., 2015), this study offers highly needed empirical evidence by exploring employees' firsthand experiences. This responds to recent calls for further qualitative, practice-oriented field research on employees' experiences with new technologies (Jacob et al., 2023; O'Neill et al., 2022). Addressing this is essential as robots are increasingly implemented across industries, but current studies have primarily been conducted in laboratories without real-world applicability (Begerowski et al., 2023). We offer insights based on actual employee experiences rather than hypothetical or imagined scenarios, making our findings particularly grounded in- and transferable to-real-world contexts.

This study contributes to the theoretical understanding of employee adaptation to working with robots, particularly in relation to human–robot cooperation in industrial environments. Consistent with career adaptability research (De Vos et al., 2020; Ployhart and Bliese, 2006; Savickas and Porfeli, 2012), our findings suggest that adaptability resources – such as curiosity or confidence-play an important role in how employees approach and cope with technological change. Employees who expressed curiosity and excitement exhibited a more adaptable disposition (e.g. were more willing to learn), while those who were skeptical displayed higher resistance. This also aligns with Schneider and Sting's (2020) findings that employees' framing of new technologies – as opportunities for exploration or as sources of threat – shapes their adaptation trajectories (Schneider and Sting, 2020). We suggest that future research continue exploring how individual differences influence employee adaptation to working with robots by, for example, clustering participants based on personal traits.

Our results further illustrate how adaptability resources shape employees' responses to the challenges they encounter during initial robot interaction. While some employees adopt adaptive strategies, like seeking help or participating in training, others use maladaptive strategies, such as ignoring the robot or switching attention to other tasks. In line with adaptability research, these differences can be explained by the presence or absence of key adaptability resources. For example, employees who are able to draw on resources such as control and confidence are more likely to adopt active coping styles and problem-focused strategies (Johnston, 2018; Ployhart and Bliese, 2006). These findings underscore the critical role of individual resources in enabling employees to adopt adaptable responses and sustain the effort required to cope with challenges in robot-augmented workplaces.

Regarding skill changes, our results indicate that working with robots requires employees to become more specialized, sometimes risking de-skilling, while simultaneously requiring the development of new skills. This aligns with research on increased skill variety due to robotization (Jonczyk et al., 2025) and with studies highlighting the tensions of simultaneously upskilling and downskilling work (Vidal, 2022). While much research has been done to outline the required skills for the future workforce (e.g. Li, 2022), our study suggests that adapting to robots involves more than just developing technical competence. It requires a cognitive shift, as employees must overcome their fears (e.g. fear of losing their jobs) and rethink how they work (e.g. adapt to new roles). Specifically, employees consistently highlighted problem-solving skills, creative thinking, openness to digitalization and willingness to learn as essential for adaptation. This suggest that even in roles that are often routine and manual in nature, the implementation of robots requires employees to exhibit high-level cognitive and behavioral skills. This contrasts with evidence that shows that, despite being among the most vulnerable to disruption from technological and structural changes, manual employees are often overlooked in strategic training investments and receive fewer opportunities for upskilling and reskilling (OECD, 2023). Further, empirical evidence shows that employees who perceive automation as a threat are less likely to engage in training (Jansen et al., 2025) as they might believe that acquiring proficiency in a wider range of skills is beyond their capacity (OECD, 2023). As a result, there is a growing mismatch between the skill demands placed on employees and the training and support they typically receive, which potentially creates labor market inefficiencies. Addressing this gap is essential, as upskilling not only enables employees to adapt to evolving work demands and maintain employability but also supports the continuous development of competencies necessary for long-term career growth, while simultaneously cultivating the personal resources – such as curiosity and confidence – that underpin career adaptability (Charalambous et al., 2015; De Vos et al., 2020).

In addition to changes in skills, the results show that adapting to working with robots also requires employees to transition to new roles. Robots shift employees' work from direct task execution to supervisory roles, affecting their jobs, responsibilities and skills. Here, adaptation is closely tied to career adaptability, as employees must adapt to changing roles and career paths (Zhang et al., 2019). This suggests that robot implementation does not just trigger skill demands but also prompts broader career-related changes, requiring employees to reorient themselves professionally. This aligns with previous research showing that technological advancements and innovations impact employees' career experiences, frequently requiring them to transition to new roles or occupations (Chen et al., 2019). Yet, these transitions also reflect tensions in how work is being reorganized. In line with prior research (Kellogg et al., 2020; Spencer, 2017), our findings show that robot implementation can create power imbalances and diminish employees' sense of autonomy and meaning. New roles focused on supervising machines may feel less empowering, as they involve less judgment and hands-on engagement. Thus, adaptation is not just about learning new skills, but about adjusting to a redefinition of one's work identity. Therefore, our findings on changing roles highlight the need to view adaptation to robotization as part of career development. We encourage future research to explore these career transitions over time, drawing on career adaptability, sustainable careers and human–robot collaboration insights.

Lastly, the study results extend the literature on employees' work-related learning behaviors and start to unpack possible ways in which employees can be supported in their learning while adapting to working with robots. While formal training (e.g. safety and technical) is essential for acquiring basic information, building confidence, reducing initial stress and minimizing the risk of accidents, much of the learning that supports employee adaptation to working with robots occurs informally by doing and from everyday interactions with colleagues. This aligns with research emphasizing that workplace learning is often embedded in routine tasks and social exchanges, where employees acquire new skills and knowledge organically by observing, experimenting and collaborating with colleagues (Cerasoli et al., 2018; Tannenbaum et al., 2010). However, informal learning opportunities are often unevenly distributed and depend heavily on workplace culture, peer availability and individual initiative, factors that can reinforce existing inequalities in who adapts effectively. Moreover, organizations tend to prioritize structured, formal training programs while often overlooking the critical role of informal learning opportunities (Kopp et al., 2021). This imbalance can limit the development of essential skills for adapting to working with robots, shifting the responsibility for learning onto employees without providing sufficient structural support. From a sustainable careers perspective, these findings underscore the importance of the context dimension, which highlights how supportive work environments – both at the work group level through constructive social support, and at the organizational level through enabling HR policies – facilitate employee adaptation (De Vos et al., 2020). Thus, integrating informal learning strategies, such as peer learning, mentorship and on-the-job experimentation, into existing training frameworks not only enhances immediate adaptation to technological change but also promotes continuous skill development and lifelong learning, helping employees sustain their careers in technology-driven environments (Bozionelos et al., 2020; De Vos et al., 2020).

This study offers several practical implications. Although the need to inform staff is well established in change management literature (Kopp et al., 2021), our results revealed that communication strategies around robot implementation frequently show deficits. Providing specific information in early phases of robot implementation helps employees understand the reasons and goals, reducing uncertainty and misunderstanding at a later stage (Wanberg and Banas, 2000). Effective communication during change is associated with positive reactions, greater acceptance and change support (Oreg et al., 2011). Beyond transparent communication, our findings highlight the value of employee involvement in the design and integration phases of robot implementation. Actively involving employees in co-designing how robots are introduced not only fosters transparency and trust but also draws on their practical expertise, thereby increasing ownership and reducing resistance (Armenakis and Bedeian, 1999).

While some scholars (e.g. Chuang, 2020) attempt to provide techniques to assist a worker in becoming “robot-proof”, no set of methods will act as a “one-size-fits-all” as adapting to working with robots is highly individual, shaped by diverse experiences, skills gaps and learning needs. We suggest that organizations seeking to implement robots conduct a thorough needs assessment prior to implementing robots. By considering employee needs during pre-implementation phases (e.g. design), organizations can enhance job quality, improve human-technology interaction and achieve higher employee satisfaction and engagement (Parker and Grote, 2022). As such, a needs assessment can serve as a tool for systematically gathering employee feedback regarding robot implementation, helping companies design tailored support and guidance that aids employee adaptation and fosters a sense of involvement.

Additionally, a needs assessment also allows organizations to identify employees' career adaptability resources and design support that strengthens these resources, enabling employees to navigate change more effectively (Savickas and Porfeli, 2012).

Finally, this study emphasizes that both formal and informal training are crucial for employee adaptation to working with robots, highlighting the importance of a blended learning approach. Results revealed key elements for training design, including addressing non-routine situations, offering guided hands-on experience, ensuring enough practice time and providing periodic sessions rather than one-time training. Despite its importance, companies frequently lack clear structures to facilitate informal learning. In addition to formal training, we recommend that organizations promote knowledge exchange by enabling less experienced employees to observe and work alongside more experienced colleagues through shadowing (Beane, 2019; Tannenbaum et al., 2010). This approach facilitates knowledge transfer, enhances peer learning and supports skills development (Ferreira et al., 2017). From a career adaptability perspective, such learning practices also help employees translate their adaptability resources into adaptive responses, fostering the development of career competencies through proactive skill development and supporting long-term career sustainability.

Despite its strengths, this study also has some limitations. First, findings are situated in the context of cooperative HRI in industrial environments, where workers and robots interdependently perform tasks in shared workspaces. Future research should examine how adaptation processes may differ in relation to other HRI types (e.g. coexistence or collaboration), as well as across industries such as healthcare. Second, because of the sampling strategy used, the employees we interviewed may have been the better-adapting ones. However, as employees openly discussed reservations and prominent issues too, we don't expect this was the case. Third, while 18 interviews were sufficient to identify recurring patterns within our specific context, the relatively homogeneous participant pool may have facilitated early saturation. Our analysis did reveal some differences based on the type of robot and field of application (e.g. differences in skills required), highlighting the need for tailored approaches to employee support during robot implementation. Conducting additional interviews, especially in other sectors such as healthcare or with different types of robotic systems, could reveal further variations and enrich the perspectives on our findings. Finally, while this study focused on employee experiences of adaptation, future research would benefit from adopting a multilevel perspective to assess how organizational strategies and practices align – or fail to align – with employees' lived experiences of working with robots.

Our study sought to understand employees' initial experiences when adapting to working with robots in industrial environments, the way in which they acquired the required knowledge and skills, and their unaddressed needs, offering insight that can inform employee-centered implementation strategies. In essence, adapting to working with robots often involves navigating ambiguity, overcoming initial challenges, developing new skills and adapting to new task routines, conditions that demand on-the-job learning, and the acquisition of knowledge through repeated exposure.

Altogether, our findings underscore the critical importance of fostering adaptable mindsets, enhancing employees' adaptability resources and creating learning environments that support both formal and informal skill acquisition. This holistic approach is vital for sustainable career development in a workplace increasingly dominated by robotics and smart technologies.

The supplementary material for this article can be found online.