Olivia Orellana 

Exploration of employee attitudes in AI adoption 

 

 

 

 

 

 

 
  

Vaasa 2025 

School of Management  
Master’s thesis in human re-

sources management 



2 

UNIVERSITY OF VAASA 
School of Management 
Author:    Olivia Orellana 
Title of the Thesis:  Exploration of employee attitudes in AI adoption 
Degree:    Master of Economic Sciences 
Programme:   Human Resources Management 
Supervisor:   Laura Urrila 
Year:    2025 Pages: 74 

ABSTRACT: 
Tämä laadullinen yksittäistapaustutkimus tarkastelee työntekijöiden asenteita liittyen tekoälyn 
käyttöönottoon organisaatioissa. Tekoälyllä viitataan monimutkaisiin teknologioihin, jotka jäljit-
televät ihmisen älykkyyttä, viitaten teknologian kognitiivisiin kyvykkyyksiin. Tekoäly mahdollis-
taa monia hyötyjä organisaatioille, sen avulla voidaan esimerkiksi optimoida prosesseja, edistää 
tehokkuutta ja innovaatiota sekä tukea päätöksentekoa. Organisaatioiden haasteeksi on kuiten-
kin muodostunut onnistunut tekoälynkäyttöönotto ja tavoiteltujen hyötyjen realisoituminen.  
 
Aikaisempi tutkimus tunnistaa työntekijöiden asenteiden merkityksen teknologisissa muutok-
sissa. Tämän voidaan todeta pätevän myös tekoälyn käyttöönottoon, joka itsessään on uniikki 
ja kompleksi organisaatiotason muutos. Työntekijöiden asenteita tarkastelemalla organisaatiot 
voivat saada käsityksen niistä tekijöistä, jotka osaltaan voivat edistää tai hidastaa tekoälyn käyt-
töönottoa.  
 
Tutkimuksen teoreettinen tausta koostuu muutosjohtamisen tutkimuksesta sekä teknologian ja 
tekoälyn käyttöönottoa käsittelevistä tutkimuksista, joiden keskiössä ovat asenteet ja ihmiset. 
Hyödyntämällä muutosjohtamista koskevaa tutkimusta mahdollistettiin yksityiskohtaisempi ku-
vaus työntekijöiden asenteista sekä niiden mahdollisista kontribuutioista tekoälyn käyttöönoton 
onnistumiselle. Valintaan vaikutti myös tekoälyä koskevan tutkimuksen vajavaisuus aihepiiristä, 
ja tällä tutkimuksella pyritään laajentamaan ymmärrystä ja tuottamaan lisää informaatiota ai-
heesta.  
 
Tutkimuksen keskeiset tulokset kiteyttävät työntekijöiden asenteiden kuvastavan avoimuutta, 
sitoutuneisuutta, kyynisyyttä sekä skeptisyyttä tekoälyn käyttöönottoa kohtaan. Näitä löydöksiä 
tukee aikaisempi muutosjohtamisen tutkimus, teknologian käyttöönoton tutkimus sekä teko-
älyn käyttöönoton tutkimus. Tutkimuksen löydöksiin perustuen organisaatioiden tulisi keskittyä 
tukemaan työntekijöiden positiivisia tekoälyn käyttöönottoon liittyviä asenteita, avoimuutta ja 
sitoutuneisuutta sekä vähentämään negatiivisia asenteita, kyynisyyttä ja skeptisyyttä, tavoitel-
lessaan onnistunutta tekoälyn käyttöönottoa. 
 
 
 
 
 

KEYWORDS: Artificial intelligence, employee attitudes, organizational change, case study 
  



3 

Table of contents 

 

1 Introduction 4 

1.1 Research background 4 

1.2 Research gap 5 

1.3 Research problem and questions 6 

1.4 Research object 7 

2 Theoretical background 9 

2.1 Artificial intelligence 9 

2.2 AI adoption 12 

2.3 Employee attitudes 16 

2.3.1 Change management literature on employee attitudes 17 

2.3.2 AI literature on employee attitudes 22 

3 Methods 27 

3.1 Case study 27 

3.2 Case subject 28 

3.3 Data gathering 29 

3.4 Analysis methods 32 

4 Findings 37 

4.1 Employee openness towards AI adoption 38 

4.2 Employee commitment towards AI adoption 41 

4.3 Employee cynicism towards AI adoption 44 

4.4 Employee skepticism towards AI adoption 48 

5 Discussion 52 

5.1 Managerial and practical implications 59 

5.2 Limitations and future research agenda 61 

5.3 Conclusions 63 

6 References 65 

  



4 

1 Introduction 

1.1 Research background 

The fast pace of change and the transformational nature of it create an environment of 

uncertainty in organizations, as the change processes are no longer linear nor planned 

years ahead following a previously planned roadmap from start to finish (Li, 2020, p. 81; 

Matsunaga, 2022). Rather they are fluid and constantly evolving, demanding constant 

informing, evaluation and updating of strategy from organizational leaders (Li, 2020, p. 

812). Organizations’ success in the world that is currently going through another tech-

nological transformation depends heavily on its ability to continuously adapt and inte-

grate new technologies, transition to new business models and to new organizational 

designs (Li, 2020, p. 814-815).  

 

Artificial intelligence (AI) is currently one of the most transformational influences affect-

ing the everyday lives of humans and organizations and its impact continues to grow 

(Morandini et al., 2023, p. 41; Lee et al., 2022, p. 1; Bankins et al., 2024a, p. 159). AI is 

commonly described as a machine’s ability of human thinking, referring to the cognitive 

abilities of such systems (Kergroach, 2017, p. 7). It is applicable for numerous purposes 

of use, such as automation of a particular tasks or parts of tasks, optimization of pro-

cesses and performance, and augmentation of human intelligence (Makarius et al., 2020; 

Plastino & Purdy, 2018; Chowdhury et al., 2022, p. 31-32; Glikson & Woolley, 2020). AI 

benefits organizations in many ways while affecting the organizations’ core processes, as 

it transforms the ways tasks are accomplished and performed, how decisions are made 

and how interactions are carried out (Deloitte, 2017, p. 20). AI in many organizations is 

used as the key source of innovation and it can provide many advantages for organiza-

tions in competitive fields (Li, 2022, p. 3; Chowdhury et al. 2022, p. 31). 

 

The transformation created by AI and other new technologies is referred to as the Fourth 

Industrial Revolution or the Second Machine Age (Jarrahi, 2018, p. 578; Im & Kim, 2022, 

p. 559). Organizations in varying fields are adopting these new technologies fast, and the 



5 

transformational influence that AI is expected to have on businesses and in the future of 

work is significant (Lee et al., 2022, p. 1; Deloitte, 2017, p. 20). However, organizations 

are struggling to gain any actual benefits and carrying out successful AI adoption pro-

cesses (Deloitte, 2017). 

 

 

1.2 Research gap 

The transformational power of AI exceeds beyond the organizations processes to its em-

ployees, and despite the form or application of the technology, AI is set to change the 

nature and meaning of work, alongside employee roles, tasks, and skills (Chowdhury et 

al., 2022; Chowdhury et al., 2023; Bankins et al., 2024b; Presbitero & Teng-Calleja, 2023). 

However, AI has been widely studied and explored from a technical standpoint as many 

studies focus on the development of technical resources and systems (Makarius et al., 

2020). The research of AI in the field of human resources management has not gained 

as much interest, even though prior industrial revolutions and technical advancements 

have proved the importance of developing human capital in the process (Jaiswal et al., 

2022, p. 1184; Makarius et al., 2020). Employees need to adapt to the transformation 

brought by AI by developing their knowledge, skills and abilities to be able to leverage 

such systems and successfully integrate them into their work (Malik et al., 2022; Pres-

bitero & Teng-Calleja, 2023; Jaiswal et al., 2022; Glikson & Woolley, 2020; Makarius et al. 

2020; Haenlein & Kaplan, 2019b, p. 23). Through successful employee adaptation organ-

izations can gain the benefits of AI adoption (Chowdhury et al., 2023; Kergroach, 2017, 

p. 7).  

 

This research is grounded in the assumption that employee attitudes act as mediators 

or variables in AI adoption processes, impacting the success of such processes (Pres-

bitero & Teng-Calleja, 2023; Malik et al., 2022; Bankins et al., 2024a). Few studies, such 

as the research conducted by Cao et al. (2021), Park et al. (2024) and Lichtenthaler (2019) 

examine this relationship between employee attitudes and AI adoption. As the previous 

AI research concerning attitudes has recognized employee attitudes contribution to the 



6 

success of AI adoption, it has usually focused on making rigid distinctions between neg-

ative and positive attitudes (Lichtentaler, 2019; Baabdullah et al., 2021; Cao et al., 2021). 

To expand the understanding of attitudes and their formation, this research aims to fur-

ther the inspection of attitudes, going deeper in the exploration into their characteristics 

and underlying meanings. This is done to fill the gap in AI research in the field of human 

resources management that currently exists by providing more information about the 

attitudinal concepts and how they are formed with regard to AI adoption in organizations. 

 

Together with AI specific literature, technology acceptance and change management 

theories concerning employee attitudes are utilized as the theoretical premises in this 

research. This is done as the introduction of AI into organizations is fundamentally a 

change process including the introduction of a new technology (Chatterjee et al., 2021; 

Bankins et al., 2024a). Leveraging change management research to AI adoption provides 

insights into the formation and characteristics of employee attitudes as well as their ef-

fect on change success in the organizational change context (Choi, 2011). Another reason 

for utilizing the different research disciplines concerns the small amount of AI research 

regarding AI attitudes in relation to the organizations efforts to adopt AI in their pro-

cesses and actions. Based on the previous research regarding AI adoption, technology 

acceptance and change management, it’s argued that managing employee attitudes is 

vital when organizations are pursuing successful adoption of AI technologies (Presbitero 

& Teng-Calleja, 2023; Malik et al., 2022; Bankins et al., 2024a).   

 

 

1.3 Research problem and questions 

As described above, a prominent need for more research of AI in the organizational con-

text and regarding the employees’ perspective exists (Im & Kim, 2022, p. 559; Chow-

dhury et al., 2022, p. 32; Suseno et al., 2022). The purpose of this study is to explore the 

employee perspective of AI adoption and expand the understanding of employee atti-

tudes towards AI adoption. The following research questions are proposed to guide and 

support the research purpose:  



7 

 

1. What attitudes do employees experience and express towards AI adoption in 

their organization? 

2. How are these attitudes formed and what characteristics and meanings do they 

portray? 

3. What is the contribution of these attitudes to the success of AI adoption in or-

ganizations? 

 

 

1.4 Research object 

The impact of AI on employees depends on multiple individual and contextual factors, 

indicating that differences in the adoption and implementation processes are likely to 

occur between employees and organizations (Bankins et al., 2024a, p. 169; Meijer et al., 

2021; Tong et al., 2021). Prior research has also pointed out that the introduction of such 

complex and cognitive technologies should be approached from a process standpoint 

including the context that it is happening in, rather than viewing it as a singular or one-

off implementation of an instrument (Meijer et al., 2021). To address the subjective and 

context specific nature of AI adoption, a case study was conducted utilizing qualitative 

methods (Yin, 2009; Piekkari & Welch, 2020).  

 

The case subject was a single organization and its frontline service employees operating 

in the retail industry in Finland. The exploration of such an organization operating in the 

retail industry is prominent, as the impact and possible business value of AI is expected 

to be significant due to its cognitive capabilities and various possible applications 

(Bonetti et al., 2022; Heins, 2023). Retail organizations adopt AI to enhance their value 

creation processes and customer experience, to gain cost reductions and performance 

advances among many other objectives (Bonetti et al., 2022). To gather the research 

data, 14 employees of the case organization participated in the research interviews. A 

data-driven analysis utilizing qualitative methods was conducted and findings were fur-

ther contrasted with theory in the discussion of this research. Main findings of the study 



8 

included the identification of four attitudinal concepts, openness, commitment, cyni-

cism and skepticism that the employees expressed towards AI adoption in the organiza-

tion.  



9 

2 Theoretical background 

2.1 Artificial intelligence 

Artificial intelligence (AI) in its earliest stages of research was contrasted with human 

intelligence by researchers like Marvin Minsky in 1968 (Haenlein & Kaplan, 2019b, p. 17). 

This formed the basis for the definition of AI as a computer’s or a machine’s ability of 

simulating human intelligence (Glikson & Woolley, 2020, p. 627; Haenlein & Kaplan, 

2019b, p. 17). Haenlein & Kaplan (2019a, p. 5) provide a more comprehensive and de-

tailed definition, as they define AI as “a system’s ability to interpret external data cor-

rectly, to learn from such data, and to use those learnings to achieve specific goals and 

tasks through flexible adaptation”. As the definition highlights, data is in the center of an 

AI system, as the system gathers knowledge and learns autonomously from it by lever-

aging approaches from machine learning (Brock & Wangenheim, 2019, p. 116; Haenlein 

& Kaplan, 2019b, p. 17-18). The ability to learn and develop continuously is considered 

one of AIs defining characteristics, as it makes AI standout from prior technology, along-

side its capability of collaboration and adaptation to human interaction (Haenlein & 

Kaplan, 2019b; Makarius et al., 2020; Lee et al., 2022, p. 2). As AI holds the ability for 

continuous and autonomous learning and development, it is applicable to perform even 

more complex and demanding tasks (Haenlein & Kaplan, 2019b).  

 

AI covers a diverse number of technologies, that are often referred to as the domains or 

sub-areas of AI. They represent the different functions of AI technologies and act as en-

ablers for more sophisticated AI systems performance (Haenlein & Kaplan, 2019b, p. 17; 

Deloitte, 2017, p. 4). According to Haenlein and Kaplan (2019a, p. 8) neural networks 

and deep learning, that is a form of machine learning, are used to generate the basis for 

most technologies considered AI. Other domains or sub-areas of AI include Internet of 

Things (IoT), robotics, computer vision, natural language processing (NLP), visual and 

speech recognition and rule-based systems (Deloitte, 2017, p. 4; Stone et al., 2022; 

Bankins et al., 2024a, p. 159). 

 



10 

Chowdhury et al. (2022, p. 33) conceptualize the applications of AI systems within busi-

ness organizations to include automated intelligence, assisted intelligence, augmented 

intelligence, and autonomous intelligence. The term automated intelligence is used to 

describe the AI system’s ability of automating a specific manual or routine task (Chow-

dhury et al., 2022, p. 33).  In a service delivery process automated intelligence can be 

leveraged in the form of a chatbot or virtual assistant (Chowdhury et al., 2022). Assisted 

intelligence refers to the systems’ ability of assisting human decision-making by provid-

ing humans with insights from data (Chowdhury et al., 2022, p. 33). Similarly with as-

sisted intelligence, augmented AI augments human decision-making while simultane-

ously learning from the interactions (Chowdhury et al., 2022, p. 33). Autonomous intel-

ligence applications are capable of autonomous development and learning, without the 

need for human intervention or guidance (Chowdhury et al., 2022, p. 33).  

 

Specific to the industry and the applications of the technology, AI brings many benefits 

to organizations (Haenlein & Kaplan, 2019b, p. 18-20). These benefits include process 

efficiency and increased productivity, enhanced analytical capabilities, support and re-

duction of human errors in decision-making and increased quality of operations (Malik 

et al., 2022, p. 337; Lee et al., 2022, p. 2; Chowdhury et al., 2022, p. 32). In more detail, 

AI can enhance productivity and improve performance through automation and optimi-

zation of business processes (Lee et al., 2022, p. 2; Deloitte, 2017). AI enabled improve-

ments in resource allocation can lead to cost reductions, as AI can take over routine tasks 

and release time for employees to work on other, more demanding tasks (Lee et al., 2022, 

p. 2). Other possible cost savings can result from improved factoring productivity and 

efficiency of processes (Brynjolfsson et al., 2019).  In addition, AI is capable of augment-

ing human decision-making and judgement, and it can reduce possible human errors and 

biases by generating important insights and producing accurate predictions (Lee et al., 

2022, p. 2; Deloitte, 2017, p. 8). AI is also used extensively as a major source of innova-

tion in many organizations and by leveraging AI applications firms can amplify their com-

petitiveness (Li, 2022, p. 3). Organizations can apply AI technologies to enhance the fea-

tures and functionality of their products and services or even create new innovations 



11 

and access new markets (Lee et al., 2022, p. 2; Deloitte, 2017, p. 4). As highlighted, a 

wide range of possible AI applications exist, but only a fraction of them have been rec-

ognized yet (Brynjolfsson et al., 2019).   

 

Considering its transformational impact on organizational life across industries and oc-

cupations, the future of work and employment seems uncertain and raises concerns 

(Bankins et al., 2024b). According to Bankins et al. (2024b, p. 2), AI influences work in 

three ways, by replacing work or work tasks, complementing or augmenting humans at 

work, or by creating new work tasks for humans. Many studies have focused on the first 

influence of AI, and the job destruction and job creation phenomena has emerged across 

industries (Huang & Rust, 2018). For example, Huang and Rust (2018) in their work in-

troduce the theory of AI job replacement addressing this phenomenon in the service 

industry. Other studies have focused on the skills and knowledge of employees, indicat-

ing that AI poses a need for human capital development (Bodea et al., 2024; Bankins et 

al., 2024b). 

 

However, in the context of human work, scholars have emphasized AIs capability of hu-

man augmentation rather than the replacement of humans (Chowdhury et al., 2022, p. 

33; Jarrahi, 2018). AI combined with human resource management can create added 

value for organizations, employees, and customers (Chowdhury et al., 2023; Deloitte, 

2017, p. 4) and as AI provides many possible applications and benefits to its usage, the 

collaborative nature of AI needs to be addressed in organizations (Malik et al., 2022). 

Chowdhury et al. (2022, p. 32) define collaborative intelligence as the effective collabo-

ration and partnership between artificial and human intelligence. According to Bankins 

et al. (2024a, p. 163) collaborative intelligence is the use of such AI systems that support 

and interact with humans to produce wanted outcomes. Jarrahi (2018, p. 583) uses the 

term human-AI symbiosis to describe the same phenomena, and states that this interac-

tive and collaborative relationship between humans and AI contributes to the develop-

ment of both sides. Bankins et al. (2024a, p. 163) state that collaborative intelligence is 

the gateway for organizations to reach the benefits of AI adoption.  



12 

 

 

2.2 AI adoption 

AI adoption is defined as the process of implementing AI enabled technologies into an 

organization and its core processes (Lee et al., 2022). An increasing number of organiza-

tions in various fields are preparing for the adoption of AI due to its many possible ben-

efits and advantages, yet still the successful integration of the technology remains a key 

challenge (Lee et al., 2022; Deloitte, 2017, p. 12). Many organizations have focused on 

the technical side of the implementation process, and heavy investments on tangible 

resources have been made by companies (Lee at al., 2022, p. 3). However, previous re-

search has shown that the actual users of the technology, that is the employees and 

other members of an organization, can have a significant impact on the success of such 

processes and therefore investing in human capital is highly necessary (Lee et al., 2022; 

Chowdhury et al., 2023, p.12). Accordingly, Chowdhury et al. (2022, p. 33) state that em-

ployees need to understand, trust, and adopt AI, so that the anticipated benefits of AI 

can be reached. Therefore, this research understands successful AI adoption as em-

ployee acceptance and behavioral intention to use the technology, going beyond just the 

implementation of AI technology or systems (Cao et al., 2021).  

 

Prior research on the topic of AI adoption has utilized varying models and frameworks 

within different research disciplines. Resource-based view (RBV), knowledge-based view 

(KVB), organizational socialization frameworks and technology acceptance models have 

been used as the theoretical premises and based on them new models and frameworks 

have been created to specifically address AI adoption (Chowdhury et al., 2022; Chatter-

jee et al., 2021). In this research we will leverage previous research on technology ac-

ceptance and AI adoption that focuses on the employee perspective of the phenomena, 

excluding the models and theories focusing on the technical side or other areas of the 

AI adoption process.  

 



13 

Technology acceptance theories such as the Technology Acceptance Model (TAM) by Da-

vis (1989) and the Unified Theory of Acceptance and Use of Technology (UTAUT) by Ven-

katesh et al. (2003) have been used previously to examine the factors influencing the 

adoption of new technologies in organizations (Trenerry et al., 2021, p. 3). Researchers 

like Cao et al. (2021), Chatterjee et al. (2021) and Lichtenthaler (2019) have later utilized 

the theories to AI adoption.  The Technology Acceptance Model (TAM) created by Davis 

(1989) recognizes the need for addressing the nontechnical side of technology imple-

mentation, referring to the actual users of the technology, for organizations to achieve 

potential benefits of it. In their work Davis (1986, p. 24) presents user attitude as a major 

determinant of actual system usage as a part of the users work. Perceived usefulness, 

that is the level to which the users see implications for usage in a particular system, and 

to what extent the users believe to benefit from its usage in their work, and perceived 

ease of use, that is  the level of effort and difficulty that the users connect to the system, 

and whether the user believes to be capable of using such system in their work effort-

lessly or with much effort, are two key variables that affect the users’ attitude, and fur-

ther the acceptance of new technologies according to TAM (Davis, 1986, s. 24). TAM2 is 

the extended version of TAM, that includes the variables of subjective norms and system-

specific technology use in the model (Trenerry et al., 2021, p. 3; Venkatesh & Davis, 2000). 

 

Following the work of Davis (1989) and other researchers in the field, Venkatesh et al. 

(2003) created the Unified Theory of Acceptance and Use of Technology model (UTAUT) 

with the intention of creating a singular unified model that explains the user acceptance 

and usage behavior of new technologies. The research conducted by Venkatesh et al. 

(2003) identified four determinants of user acceptance and usage behavior to include 

performance expectancy, effort expectancy, social influence, and facilitating conditions. 

In addition, four moderators were identified to include age, gender, experience and vol-

untariness (Venkatesh et al., 2003). The moderators affected how strong the influence 

of each determinant was (Venkatesh et al., 2003). In the model performance expectancy 

describes the employees’ belief of possible performance gains resulting from systems 

usage and it is considered the strongest indicator of user intention. Effort expectancy 



14 

includes the effort and ease of use that the system requires (Venkatesh et al., 2003, p. 

450). Social influence contains the individual perception of importance that others be-

lieve that they can use the systems (p. 451-452). Facilitating conditions is the individuals’ 

perception of the organizational and technical infrastructures, and whether they believe 

that they are capable of supporting and facilitating the use of the system (Venkatesh et 

al., 2003, p. 453-454). The research conducted by Venkatesh et al. (2003) managed to 

connect the determinants to intention to use and further to actual usage behavior, 

providing insights on the important factors influencing the process of technology ac-

ceptance. 

 

Researchers like Makarius et al. (2020) and Bonetti et al. (2023) have created specific 

models and frameworks considering the adoption of AI in organizations, addressing the 

unique and complex nature of the technology and its adoption. Makarius et al. (2020) 

studied the social element of AI and utilized Socio Technical Systems (STS) theory and 

organizational socialization approach to form the AI socialization framework. The AI so-

cialization framework comprises of three phases, employee anticipation phase, AI-Em-

ployee encounter phase, and symbiotic metamorphosis phase (Makarius et al., 2020, p. 

267). The employee anticipation phase happens before AI has yet to be brought into an 

organization (Makarius et al., 2020, p. 267). The phase consists of sensing employees’ 

attitudes and uncertainties towards AI, addressing possible challenges and informing 

employees of the coming change and its implications (Makarius et al., 2020, p. 267). The 

AI-employee encounter phase occurs when AI systems are introduced into the organiza-

tion and employees start to socialize with the AI systems (Makarius et al., 2020, p. 267-

268). Symbiotic metamorphosis phase occurs after the successful integration of AI, and 

organizations can then expect to realize the benefits of AI adoption (Makarius et al., 2020, 

p. 269). Makarius et al. (2020) consider employee type and technological readiness to 

impact the phases, as employees may hold varying expectations and experiences regard-

ing AI that influence how AI is encountered (Makarius et al., 2020, p. 267-268). Techno-

logical readiness refers to the organizations’ capabilities of utilizing its technological as-

sets, and higher levels of technological readiness can prepare organizations and 



15 

employees for the integration of a new AI technology (Parasuraman, 2000; Makarius et 

al., 2020, p. 269). Technological readiness may also moderate the expectations and atti-

tudes that employees have towards AI, contributing to the anticipation and encounter 

phases (Makarius et al., 2020, p. 267-269). Sociotechnical capital that is according to 

Makarius et al. (2020, p. 265) “the combination of AI technology and people in organi-

zations that leads to a source of competitive advantage for an organization” is the pur-

sued outcome of a successful AI adoption process. 

 

Bonetti et al. (2023) created the Practice co-evolution model to address AI adoption in 

retail practices. According to Bonetti et al. (2023, p. 867) practice co-evolution is defined 

as “the process that undergirds successful AI integration and enables retail employees’ 

sustained usage of AI” and it includes the needed transformation, improvement, and 

modification of employee practices to successfully collaborate with the technology (p. 

869). Practice co-evolution manifests in through three courses of action, co-envisioning, 

co-adaptation and co-(re)aligning (Bonetti et al., 2023, p. 870). In co-envisioning, the ob-

jectives of AI adoption are outlined and the action needed to reach them is planned 

accordingly. In co-adapting, the co-envisioned plans of action are adopted to usage in 

the given context, and they are modified to better suit it through co-(re)alignment ac-

tions (Bonetti et al., 2023, p. 870). Practice co-evolution is fostered through practice en-

ablement, that is the knowledge possessed, shared and utilized by the employees taking 

part in the co-evolution process (Bonetti et al., 2023, p. 870). According to Bonetti et al. 

(2023) successful AI implementation relies on collaboration and the involvement of par-

ticipants throughout the phases of co- envisioning, co-adaptation and co-(re)aligning.  

 

Baabdullah et al. (2021) in their research investigating AI acceptance in B2B SMEs con-

sider AI enablers, that are the driving forces promoting the adoption of AI technology. AI 

enablers include technology road mapping, professional expertise, and attitude 

(Baabdullah et al., 2021, p. 259-260; Shon & Kwon 2020). Technology road mapping 

stands for the strategic planning process of new technology implementations, including 

the research, development, and implementation of the technologies (Baabdullah et al., 



16 

2021, p. 259-260). Professional expertise includes the employees’ knowledge, skills, and 

experience of AI systems (Baabdullah et al., 2021, p. 260). Attitudes in AI adoption stand 

for the negative or positive feelings employees hold against AI (Baabdullah et al., 2021, 

p. 260; Cao et al., 2021). Baabdullah et al. (2021) found that technology road mapping 

and attitudes were significant to AI adoption success in organizations, disregarding pro-

fessional expertise. Along with the contributing AI enablers, organizations’ AI readiness, 

that is the organizations’ abilities to support AI practices including technology infrastruc-

ture, awareness, and technicality, was concluded to significantly influence the ac-

ceptance of AI (Baabdullah et al., 2021). 

 

As shown, the discussed technology and AI adoption research acknowledges the contri-

bution of employee attitudes regarding the success of such organizational changes (Da-

vis, 1989; Venkatesh et al., 2003; Baabdullah et al., 2021; Makarius et al., 2020). However, 

the research does not conceptualize or define the specific characteristics of employee 

attitudes nor explore their specific contributions in such processes. To further explore 

the meanings and importance of employee attitudes, this research now defines em-

ployee attitudes in general and utilizes change management research to provide a better 

understanding of their contribution to change processes in organizations. The few AI 

studies providing more thorough explorations and conceptualizations of employee atti-

tudes are also covered in the following sections of this research. 

 

 

2.3 Employee attitudes  

Eagly and Chaiken (1993, p. 1) define attitude “a psychological tendency that is expressed 

by evaluating a particular entity with some degree of favor or disfavor”. According to 

Eagly and Chaiken (2007) attitudes construe from three dimensions; tendency, entity 

(attitude object), and evaluation. Attitudes are created as an object is identified by the 

individual, and the object that can be in the form of concrete or abstract (Eagly & Chaiken, 

2007, p. 583-584). The attitude object or entity is the target of evaluation carried out by 

individuals, and it can be “overt or covert, or cognitive, affective, or behavioral” (p. 583). 



17 

Eagly and Chaiken (2007) continue by making a distinction between experiencing atti-

tudes and expression of attitudes. They highlight that attitudes are inner tendencies of 

individuals, and expressions are the manifestations of attitudes (p. 586). Past experi-

ences affect the formation of either positive or negative tendency to respond (Eagly & 

Chaiken, 2007). They state that attitudes “can be formed or expressed through cognitive, 

affective, and behavioral processes” (Eagly & Chaiken, 2007, p. 592). Eagly and Chaiken 

(2007, p. 593) also recognize that attitudes can manifest as either explicit or implicit. 

Explicit attitudes are attitudes that the individual is aware of experiencing, and implicit 

attitudes are those that individuals experience unconsciously (Eagly & Chaiken, 2007, p. 

593). Attitudes have been identified to guide human behavior (Glasman & Albarracin, 

2006). 

 

Following this definition of attitudes provided by Eagle and Chaiken (1993;2007) this re-

search now aims to provide a comprehensive understanding of employee attitudes in 

changes and their contribution to AI adoption. Firstly, attitudes are discussed in this re-

search from a change management perspective. This is done as change management 

research has contributed heavily to the research of employee attitudes in organizations 

and therefore provides a strong theoretical basis for this research. Another reason for 

this regards the small amount of AI research considering employee attitudes and their 

contribution to the success of AI adoption. Following the exploration of change manage-

ment literature on employee attitudes, this research aims to cover the AI specific re-

search that considers employee attitudes in AI adoption. This aims to provide insight to 

how AI researchers have defined and discussed employee attitudes, and how they are 

connected to the success of AI adoption in organizations. 

 

 

2.3.1 Change management literature on employee attitudes 

Organizational change is mandatory for organizations’ success and survival (Matsunaga, 

2022, p. 118). As organizations in their core are comprised of humans, the importance 

of those humans and their commitment to, support and acceptance of change is crucial 



18 

for successful organizational changes (Hechanova et al., 2018, p. 915; Choi, 2011). This 

individual centered approach has been recognized in previous research, as it’s been 

stated that many change processes fail to succeed as change leaders have overlooked 

the importance of individual employees or as Al-Haddad and Kotnour (2015, p. 244) refer 

to it the personal dimension of change (Choi, 2011, p. 479). With this understanding of 

individual employees’ importance in change processes, the importance of those employ-

ees’ attitudes has proved its importance in previous research (Miller et al., 1994 & Choi, 

2011). Change management can provide solutions to navigate organizational changes 

successfully, and it can be helpful when addressing the reactions and attitudes of em-

ployees in organizations (Hechanova, et al., 2018, p. 916). Change management is de-

fined by Moran and Brightman (2001) as “the process of continually renewing an organ-

ization’s direction, structure, and capabilities to serve the ever-changing needs of exter-

nal and internal customers”. 

 

Choi (2011) in their work present four key attitudinal constructs that are frequently pre-

sent in research concerning employee attitudes in change.  The constructs are readiness 

for change, commitment to change, openness to change and cynicism about organiza-

tional change. Other important emotions and perceptions relating to changes include 

skepticism, trust and anxiety (Stanley et al., 2005; Miller et al., 1994). The attitudinal 

constructs do not form in a void, but they are rather comprised of multiple factors (Choi, 

2011). As managers cannot control what attitudes the employees are displaying, they 

can influence the factors contributing to those, such as the beliefs and values of employ-

ees (Oakland & Tanner, 2007, p. 17). Miller et al. (1994, p. 61) refer to antecedents, that 

are “the factors which influence employees’ evaluation of whether the change should 

be supported, viewed with indifference, or opposed”.  

 

According to Armenakis et al. (1993) readiness for change contributes to the effective-

ness and success of organizational change efforts. Readiness includes the individual’s 

beliefs and perceptions of the need for change and the individual and collective abilities 

to overcome changes (Armenakis et al. 1993; Eby et al., 2000). Readiness is fostered by 



19 

communication and information sharing that justifies and provides support of the need 

for change (Armenakis et al., 1993). Building trust and confidence in the individual and 

collective abilities to overcome the change is also crucial, and it can be carried out 

through information sharing (Armenakis et al., 1993). Alongside trust, organizations 

should provide support and actively participate employees in the change (Armenakis et 

al., 1993; Eby et al., 2000). Readiness contributes to either resistance or support of 

change efforts, as building readiness disseminates resistance (Armenakis et al., 1993).  

 

Openness to change is defined by Wanberg and Banas (2000) as the acceptance and 

positive view of changes, and it forms the basis for successful organizational changes 

(Miller et al., 1994). Openness includes the individual’s willingness to cooperate and sup-

port the change and a positive perception of possible benefits resulting from the change 

(Miller et al., 1994). Certain individual characteristics have been connected to employee 

openness, such as cognitive and behavioral flexibility, curiosity and openness to experi-

ence, and need for achievement (Miller et al., 1994; Choi, 2011). Factors fostering open-

ness include information sharing and change specific self-efficacy (Wanberg & Banas, 

2000; Miller et al., 1994). Openness represents the opposite of resistance to change 

(Wanberg & Banas, 2000; Miller et al., 1994). 

 

Commitment to change is defined by Herscovitch and Meyer (2001, p. 303) as the psy-

chological mindset binding an individual to act in favor of a change initiative. It repre-

sents the individuals’ willingness to participate and support the change, indicating a 

more positive affect (Fedor et al., 2006). Commitment construes of affective, normative 

and continuance commitment (Herscovitch & Meyer, 2001, p. 303). Affective commit-

ment forms as an individual desires to take part in the change and they believe in the 

possible benefits that the change may bring. The individual may also sense an internal 

urge to act in favor of the change initiative (Herscovitch & Meyer, 2001, p. 320). Individ-

uals expressing continuance commitment recognize possible costs with failure to sup-

port the change initiative and therefore feel like they don’t have a choice in whether to 

do so or not (Herscovitch & Meyer, 2001, p. 320). Individuals experiencing normative 



20 

commitment feel a strong obligation and duty to work in favor of the change (Her-

scovitch & Meyer, 2001, p. 321). Commitment to change forms from factors including 

the nature of the change, its perceived fit with the organization, and provided infor-

mation regarding the change (Choi, 2011; Shum et al., 2008). Other factors independent 

from the actual change such as organizational culture and the quality of relationships 

inside the organization, previous experience from similar situations and the employees’ 

level of change self-efficacy also impact commitment (Shum et al., 2008, p. 1358-1359; 

Choi, 2011; Herold et al., 2007, p. 948). Job motivation and job satisfaction can also affect 

the employees’ commitment to change (Parish et al., 2008). 

 

Cynicism about organizational change is considered a negative attitude (Stanley et al., 

2005). It is connected to organizational cynicism, that refers to the employees’ beliefs 

and experiences of unfair actions and distrust, fraudulence or insincerity and negative 

affect toward the organization (Dean et al., 1998, p. 345; Choi, 2011). Andersson (1996) 

uses the term employee cynicism, that is defined as “an attitude characterized by frus-

tration, hopelessness, and disillusionment, as well as contempt toward and distrust of 

business organizations, executives, and/or other objects in the workplace” (p. 1395). 

Cynicism can lead to employee behaviors that are against the organizations’ benefit, and 

harm the employee itself as cynicism has been linked to many negative employee out-

comes such as burnout and emotional fatigue (Choi, 2011; Dean et al., 1998, p. 345). In 

the organizational change context cynicism can appear as employees distrust in decision-

makers and questioning of the motives and reasonings of the change and their fairness 

(Stanley et al., 2005). These can be based on employees’ previous experience from failed 

change attempts or evaluations of decision-makers liability (Stanley et al., 2005). Em-

ployees experiencing cynicism towards change also usually blame the management and 

other organizational leaders of their negative reaction, and varying personality and situ-

ational factors impact the development of cynicism towards organizational change (Dean 

et al., 1998, p. 344; Choi, 2011). Closely related to cynicism, the attitude of skepticism is 

brought up or the two attitudes are considered as one unified attitudinal construct (Stan-

ley et al., 2005). Taken apart from cynicism, skepticism towards organizational change 



21 

includes the questioning of the effectiveness and success of change efforts (Stanley et 

al., 2005). 

 

According to Choi (2011, p. 480) each attitudinal construct reflects the individual’s judge-

ment of the specific change initiative, and acts as predictors of acceptance or resistance 

towards the change. Successful change processes require employees that are motivated 

and committed to the change and willing to co-operate, or in other words acceptive of 

the change (Proctor & Doukakis, 2003, p. 268; Miller et al., 1994). Commitment, open-

ness and readiness are characterized as the more positive attitudes that promote change 

acceptance (Herold et al., 2007; Wanberg & Banas, 2000; Fedor et al., 2006; Choi, 2011). 

In contrast, cynicism and skepticism have been recognized by previous research to lead 

to resistance (Stanley et al., 2005). Resistance in change is common in organizations, as 

it can arise from employees feeling uncertain and fearing the unknown, they lack infor-

mation about the change, they feel like the change may pose threats to their status, they 

fear failure or have trouble identifying possible benefits of the change (Proctor & Douka-

kis, 2003, p. 268). Consideration and management of resistance is crucial for change suc-

cess, as it can bring costs and even delay changes (Pardo del Val & Martinez Fuentes, 

2003, p. 148). Pardo del Val and Martinez Fuentes (2003, p. 149-151) identified the 

sources of resistance to include deeply rooted values, relating to political and cultural 

deadlocks inside the organization and its practices, and lack of necessary capabilities to 

accommodate the change process successfully. Other factors included unclarity in the 

need for change, lack of information sharing, differences in motivation and interest 

among employees and management, perceived fit between the change and the organi-

zation and felt cynicism (Pardo del Val & Martinez Fuentes, 2003, p. 149-151). They con-

tinue by stating that the transformational impact of change affects the level of resistance, 

indicating that the more disruptive the change is, the more resistance can be anticipated. 

 

 



22 

2.3.2 AI literature on employee attitudes 

Previous research considering attitudes and AI has usually refers to attitudes in general, 

making distinctions between positive and negative attitudes. Dwidedi et al. (2017, p. 213) 

in their work considering technology adoption define attitudes as "an individual’s posi-

tive or negative feelings about performing a target behavior". Utilizing this definition 

provided by Dwidedi et al (2017) in the AI adoption context, attitudes can be understood 

as an individual’s positive or negative feelings about using AI.  

 

Consideration of both the positive and negative attitudes is important as they have a 

major impact on acceptance of new technologies (Lichtentaler, 2019). In addition, Gur-

soy and Huang (2024) and Tong et al. (2021, p. 1625) make the notion that differences 

between employees and how they perceive AI should be addressed accordingly, as dif-

ferences between organizations employees is likely to occur in the ways they react to the 

change and how they are able to benefit the implementation process of AI in organiza-

tions. By acknowledging and addressing the differences in the workforce, organizations 

can gain the benefits of AI adoption (Tong et al., 2021, p. 1625). Lichenthaler (2019, p. 

41) also recognizes this need for active management of employee attitudes, as regard-

less of the attitudes positive or negative characteristic, they can still have negative effects 

on organizations change initiatives considering technological change. In addition, it has 

been stated that employees can experience both positive and negative attitudes simul-

taneously and exploring the root causes generating these attitudes and reactions in em-

ployees can further aid the AI adoption process (Bankins et al., 2024a, p. 167; Lichten-

thaler, 2019). To better understand how attitudes are described and utilized in AI re-

search, this research provides a brief outlook into previous research of the topic includ-

ing the inspection of attitudes and their descriptions as well as the factors contributing 

to the formation of employee attitudes in AI adoption. 

 

Cao et al. (2021) in their work examined middle and senior managers’ attitudes towards 

AI and its impact on AI usage in decision-making. To describe and define the factors in-

fluencing the formation of attitudes and behavioral intentions, Cao et al. (2021) 



23 

developed the Integrated AI Acceptance-Avoidance model (IAAAM). The model consid-

ers attitudes as either positive or negative in their nature, that further contribute to ei-

ther acceptance or avoidance behaviors. According to Cao et al. (2021) positive attitudes 

foster acceptance behavior (usage of AI systems) and negative attitudes foster avoidance 

behavior (disuse of AI systems). It was concluded in the research that expected perfor-

mance gains and ease of use relating to the AI systems positively influenced attitudes. In 

contrast, personal well-being concerns relating to felt stress and anxiety and perceived 

threats the managers connected to AI negatively affected attitudes (Cao et al., 2021).   

 

Lichtenthaler (2019) in their work provides a more detailed examination of the positive 

and negative employee attitudes and their effect on AI adoption. Lichtentaler (2019) de-

fines intelligent automation -attitudes (IA) as the more positive attitudes that employees 

can hold towards AI. The employees expressing IA attitudes are more open to using AI 

solutions, as they hold a stronger emphasis on rational decision-making, convenience 

and optimization, and are able to comprehend the possible benefits that the AI systems 

could bring to their work (Lichenthaler, 2019, p. 41). The more negative attitudes are 

defined as no human interaction -attitudes (NHI) (Lichtenthaler, 2019). According to 

Lichtentaler (2019) the employees displaying NHI attitudes value and favor human inter-

action and characteristics and express limited openness to using AI solutions regardless 

of their usability (Lichenthaler, 2019, p. 40). Lower emphasis is put on convenience, effi-

ciency and optimization of work processes, as the fears and concerns regarding the use 

of AI override the possible positive effects (Lichenthaler, 2019, p. 40-41). The negative 

attitudes employees experience are usually connected to fears and threats that the em-

ployees associate with AI (Bankins et al., 2024a; Lichenthaler, 2019). The uncertainties 

regarding how AI will affect the individual, their work and the organization can foster 

negative attitudes (Lichenthaler, 2019; Bankins et al., 2024a).  

 

To better understand the formation of employee attitudes towards AI, Lichtenthaler 

(2019) lists five categories of factors that include personal characteristics such as open-

ness and affinity towards new technology (1), the AI technology itself and its perceived 



24 

benefits and the ease or effort of use (2), voluntary or required use of AI (3), employee 

experience of using AI systems (4), and the environment (5). According to them, employ-

ees who are more open, technology-driven and convenience oriented are more likely to 

experience positive attitudes towards the introduction of new technology (Lichtenthaler, 

2019, p. 42). Following TAM created by Davis (1989), Lichtenthaler (2019, p. 42) indicates 

that negative attitudes will be reduced if the employee perceives a potential benefit 

gained from using the AI application or system, and when the usage of the system does 

not require much effort. In addition, Lichtenthaler (2019, p. 42-43) states that the incen-

tive and frequency of the interaction between humans and AI influences attitudes, as 

voluntary and routine-based use of technology positively affects attitudes, and vice versa. 

Little experience using AI systems or applications is likely to result in more negative atti-

tudes but as employees gain experience from AI systems and applications, they are likely 

to go through a change in their attitudes from negative to more positive (Lichtenthaler, 

2019, p. 43). Standardization of behavior is also a major contributor, as new behavior 

such as the usage of a new technology, becomes standard, fewer negative attitudes are 

reported (Lichtenthaler, 2019, p. 43). Incentives from inside and outside the organization 

affect the formation of attitudes alongside the social factor of AI that relates to the de-

sirability to use the system and whether it is socially encouraged and accepted to do so 

(Lichtenthaler, 2019, p. 43).  

 

In line with the work of Lichtentaler (2019), Gursoy and Huang (2024) in their work ad-

dress the contradictory influence of AI integration on service employees’ attitudes and 

examine their influence on employee service behavior. They propose two mechanisms 

that affect the formation of employee attitudes and reactions to AI integration by lever-

aging the transactional theory of stress by Lazarus and Folkman (1981). Following the 

theory, Gursoy and Huang (2024) consider the integration of AI as either a challenge or 

hindrance stressor for employees that is affected by individual and situational factors. 

Hindrance stressors are considered to have a more negative influence, and challenge 

stressors are perceived as more positive stressors, promoting motivation, development 

and achievement among employees (Gursoy & Huang, 2024, p. 2). What they found was 



25 

that employees who considered the integration of AI as a challenge experienced more 

positive outcomes and exhibited enhanced service behavior.  In contrast, the employees 

who experienced the same situation as a hindrance had experienced increased job inse-

curity and their proactive service behavior declined (Gursoy & Huang, 2024, p. 6). Gursoy 

and Huang (2024) propose IT-mindfulness, that is according to Thatcher et al. (2018, p. 

832-834) “a dynamic IT-specific trait, evident when working with IT, whereby the user 

focuses on the present, pays attention to detail, exhibits a willingness to consider other 

users, and expresses genuine interest in investigating IT features and failures”, as a me-

diating factor contributing to the formation of the stressors. 

 

Bankins et al. (2024b) in their study researched AIs impact on work tasks, employees’ 

responses to changes and AIs effect on future work skills. Bankins et al. (2024b) propose 

that examining the influences that AI adoption has on employees, both positive and neg-

ative, can help to better understand the factors formulating the responses among em-

ployees. They recognize the varying responses that employees display relating to AI 

adoption as fear of replacement, positive and negative expectations of AIs impact and 

its applicability, and excitement and curiosity. Positive influences that employees per-

ceive of AI, like AI enabled positive changes to work tasks and job design and perceived 

quality of the AI systems, promoted more positive reactions and attitudes towards AI 

(Bankins et al., 2024b, p. 3). Vice versa, the anticipated negative impact of AI on employ-

ees and their work, like increased controlling and monitoring of employees, affected the 

formation of more negative reactions amongst employees (Bankins et al., 2024b, p. 3). 

 

Park et al. (2024) in their research concerning AI and attitudes considers AI specific atti-

tudes in addition to general attitudes towards new technologies. According to Park et al. 

(2024) this inspection of AI specific attitudes enables more insightful information about 

the attitudes and their contributions to employee attitudes towards AI. The perceived 

humanlikeness of AI that includes the extent to which AI applications are perceived to 

possess human-like characteristics, and the perceived adaptability that includes the ex-

pected learning and adaptability capabilities of AI, capture the AI specific characteristics 



26 

of attitudes (Park et al., 2024, p. 924-925). Other attitudes included the quality of AI, AI 

use anxiety, job insecurity related to AI and personal utility (Park et al., 2024). According 

to Park et al. (2024) together the attitudes form a comprehensive description of em-

ployee attitudes towards AI in organizations. Further, the attitudes predict workers’ in-

tention to use AI and work outcomes (Park et al., 2024).  



27 

3 Methods 

3.1 Case study 

This is a qualitative case study that explores how employees experience and perceive AI 

adoption with the focus of conceptualizing employee attitudes. The research investi-

gates a single organization and its employees in their natural environment. The case 

study method was chosen as it enables a thorough investigation of the case subject and 

presents the opportunity for exploring and understanding the complex social phenom-

ena that is connected to the topic of research (Yin, 2009; Makarius et al., 2020). In addi-

tion, context specific characteristics and individual factors can be explored and under-

stood utilizing the method in the research (Piekkari & Welch, 2020; Yin, 2009).  

 

Yin (2009, p. 18) defines a case study as an “empirical inquiry that investigates a contem-

porary phenomenon in depth”. Case studies are usually qualitative studies, although re-

searchers can use quantitative methods or data as a part of a case study (Piekkari & 

Welch, 2020). Qualitative case studies utilize a large amount of evidence, that can be in 

the form of speech, texts, images, observations etc. (Piekkari & Welch, 2020). The evi-

dence can be gathered through interviews, inquiries, documents, or by observations 

conducted by researchers (Yin, 2009). A case study can include the investigation of a 

single case or multiple cases, and the case that is the subject of the study can be an 

individual, a group, an institution, or a community (Gillham, 2010; Piekkari & Welch, 

2020; Yin, 2009). The investigation is conducted within the natural context of the case 

(Piekkari & Welch, 2020; Yin, 2009, p. 18). Defining the context in case studies is im-

portant to enlighten the boundaries between it and the phenomenon (Yin, 2009, p. 18; 

Piekkari & Welch, 2020). Case studies aim to explain, explore, or describe a phenomenon, 

while answering how and why questions (Yin, 2009). The goal of conducting a case study 

is to generate additions to knowledge that can be further generalized to theoretical 

propositions (Yin, 2009, p. 15; Gillham, 2010). As Yin (2009) specifies, case studies are 

not meant to produce directly generalizable proposition to populations or universes 



28 

hence the value of a case study can be measured by its ability to expand and generalize 

theories. 

 

According to Yin (2009) and Piekkari and Welch (2020) conducting single case study is a 

justified research method for certain types of case studies, including critical cases, reve-

latory cases, common cases and extreme or unusual cases. A critical case study tests 

previously presented and established theories in a specific context, whereas revelatory 

case studies examine cases that have not yet been studied by the researcher or research-

ers (Yin, 2009; Piekkari & Welch, 2020). Common case studies examine a typical case and 

in contrast, extreme or unusual case studies examine rare or exceptional occurrences 

(Yin, 2009; Piekkari & Welch, 2020).  

 

 

3.2 Case subject 

The case subject of this research is a single organization that sells and provides goods to 

consumers in Finland. The organization operates in several hundred locations in the 

country, producing tens of millions of customer interactions yearly. The organization also 

operates internationally, as they take part in business with product manufacturers and 

suppliers around the world. In 2023, the company’s turnover was just over a billion euros, 

and their profit was around 40 million euros. Among their retail activities, the organiza-

tion holds a specific societal purpose taking part in supporting the health and well-being 

of the Finnish society. The organization does not aim to maximize revenues or profits 

from their business activities, as the organization’s primary goal is related to their socie-

tal purpose as a responsible retailer of consumer goods. The company strategy consists 

of four pillars, including a promise of high-quality service, inclusivity, sustainability and 

responsibility. Among responsibility, transparency in activities with stakeholders is high-

lighted as a defining characteristic. 

 

The organization’s actions are highly supervised and governed and a well-defined admin-

istrative order is present in the organization. The administrative power and influence are 



29 

distributed in the organization between different units, that includes an administrative 

board, organizational board and its committees, CEO and the executive team. The organ-

ization employs a couple of thousand people on average, and most of which are em-

ployed as salespeople. The organizational culture consists of feelings of solidarity, clarity 

and efficiency and it is described to result from supportive management and developing 

one’s own competencies alongside knowledge sharing among the employees. Inclusion, 

acceptance, and diversity are described as defining constructs of the organization’s staff, 

as it comprises of employees from different demographics such as age, sexual orienta-

tion, cultural background and religion.  

 

The organization values highly the quality of customer service in their actions and as a 

way to create business value and competitive advantage. This translates to their invest-

ments in training and continuous efforts to support employee development throughout 

their careers. On-the-job learning is the main practice of competence development, and 

digital tools are utilized in the process. The competence of employees in the organization 

comprises of product knowledge, knowledge sharing and high-quality customer service. 

The payroll system in the organization is knowledge-based, and it offers opportunities 

for employees to influence their pay by developing their personal competence. The com-

pany holds a good reputation as an employer and has even received local acknowledge-

ment as a result. 

 

 

3.3 Data gathering 

The data for the research was gathered through qualitative thematic interviews. Inter-

views as a data gathering method are commonly used in various fields of research, as 

they are applicable to many research purposes (Hirsjärvi & Hurme, 2022). Interviews as 

a data gathering method refer to the process of gathering opinions, thoughts and views 

of humans, or the interviewees, in verbal form about the study subject (Hirsjärvi & 

Hurme, 2022). In an interview two or more people as in the interviewer and the inter-

viewees meet each other to have an interaction about the study subject. During the 



30 

interview, the interviewee paints a picture of their reality and how they perceive it, and 

that is communicated to the interviewee with words and expressions (Hirsjärvi & Hurme, 

2022). Interviews in their core are social interactions between two or more people, and 

many factors contribute to their success as a method (Hirsjärvi & Hurme, 2022). It is 

important to consider the entirety of the research and their contribution to it, and the 

value of definitions, meanings and factors relating to language and linguistic expressions 

(Hirsjärvi & Hurme, 2022).  

 

The interviews can have differing forms, meaning that they can be open, semi-structured 

or structured in their character. Hirsjärvi and Hurme (2022) introduce the definition of 

thematic interviews in their work. Thematic interviews as an interview form are placed 

between structured interviews and unstructured interviews, and they are sometimes re-

ferred to as semi-structured interviews (Hirsjärvi & Hurme, 2022). The defining character 

of thematic interview is that the direction or focus of conversation is directed to certain 

subjects by the interviewee, rather than certain specific interview questions (Hirsjärvi & 

Hurme, 2022). The subjects are defined beforehand by the interviewee and the pre-

sumption in thematic interviews is that the experiences, thoughts, beliefs and emotions 

of an individual can be examined when leveraging this method (Hirsjärvi & Hurme, 2022).  

 

Interviews were chosen as the method for gathering the research material, as it gives 

the opportunity to bring out underlying motives while enabling flexibility and adaptation 

during the procedure, as well as support for the human centered perspective of the 

study (Hirsjärvi & Hurme, 2022). The interviews conducted in this research were semi-

structured thematic interviews between the interviewer and a single interviewee. The 

interviewee had selected certain themes for the interviews that guided the direction of 

the interview, but no specific interview questions were defined beforehand. The inten-

tion was to carry out interactions with the interviewees that would generate deep and 

valuable information for the research (Hirsjärvi & Hurme, 2022). It was predicted that 

the subject of the research may emerge strong emotional reactions from employees as 

well as varying opinions and meanings underlying them. The interviews enabled 



31 

expression and capturing of these emotions and reactions to the subject, that would 

have otherwise been left uncaptured if other data gathering methods were utilized 

(Hirsjärvi & Hurme, 2022).  The aim of the interviews was to generate open answers, and 

the interviewees were encouraged to answer the questions with their own words, de-

scribing their own experiences, emotions, expectations and possible prejudices that they 

had about the study subject. Thematic interviews as a data gathering method enabled 

this, as the importance of individual’s interpretations and meanings are in the center of 

the method (Hirsjärvi & Hurme, 2022). To enable the capturing of possible sensible em-

ployee experiences and reactions the participants were promised anonymity (Gioia et 

al., 2012). In addition to conduct ethical research, the participants were informed about 

the research proceedings before signing their consent to take part in the research.  

 

A cross-sectional study design was utilized to gather the evidence for the study. The in-

terviews were conducted in direct interaction between the interviewee and the inter-

viewer during the months of May and June of 2024, and the duration of the interviews 

varied between 40 minutes to 1.5 hours. The interviews were recorded and later tran-

scribed into written text. It was planned that the interviews included 10 to 15 people 

from inside the case organization and eventually 14 participants were interviewed. It was 

specified that the participants consisted of sales personnel, excluding managers and 

other job positions inside the organization. Participation in the interviews was voluntary, 

and interest to participate in the interviews was requested and selection of participants 

was conducted based on the responses. The large number of willing participants made 

it possible to select the interviewees to best serve the intention of the study. The selec-

tion of interviewees was made to generate a data set that reflected the organization 

comprehensively, as employees representing both genders and varying age groups, ed-

ucational backgrounds and varying tenure in the organization were selected. 12 of the 

research participants were female, and the remaining 2 males. The participants will be 

later referred to as Participant 1, Participant 2 and so on based on the order of interviews 

conducted. 

  



32 

  

3.4 Analysis methods 

After the data was gathered, a qualitative analysis was conducted. A qualitative analysis 

concentrates on the content of the data (Gunther & Hasanen, 2021).  By analytically exa-

mining the content interpretations can be made about it, and those interpretations can 

be further inspected by contrasting them with theory. A major part of qualitative analysis 

is the researchers own thinking and its contribution to it (Gunther & Hasanen, 2021).  A 

data set itself does not give ready answers for research questions, as it is the analysis 

that produces the insights to the subject and more information about the research topic 

from a chosen viewpoint (Gunther & Hasanen, 2021). The aim of qualitative analysis is 

to refine the data set to then be able to examine and draw conclusions about the content 

in relation to the research questions (Gunther & Hasanen, 2021). 

 

Different forms of qualitative analysis utilize and position theory differently (Saaranen-

Kauppinen & Puusniekka, 2006). In data-driven or inductive analysis the formation of 

theory is the aim of the analysis, and the data set is the starting point for the process 

excluding previous views, opinions and conclusions from it (Saaranen-Kauppinen & 

Puusniekka, 2006; Tuomi & Sarajärvi, 2018). No units of analysis are defined beforehand, 

as the data determines what is the focus of the analysis (Saaranen-Kauppinen & 

Puusniekka, 2006). When utilizing theory-driven or deductive analysis, the analysis is 

based on existing theory or theoretical model (Saaranen-Kauppinen & Puusniekka, 2006). 

The analysis includes testing the chosen theory or theories in contrast to the research 

data and research topic (Saaranen-Kauppinen & Puusniekka, 2006). This means that the 

units of analysis are defined in advance (Saaranen-Kauppinen & Puusniekka, 2006). The-

ory-directed analysis is usually located between inductive and deductive forms of analy-

sis, as it combines the two (Tuomi & Sarajärvi, 2018). In theory-directed analysis, theory 

can aid the analysis by guiding it based on previous information on the topic, but no units 

of analysis are determined in advance (Tuomi & Sarajärvi, 2018). The aim is not to test 

theory or theoretical models, but rather to create new insights to the topics previously 

presented in research (Tuomi & Sarajärvi, 2018). 



33 

 

Different methods for data analysis, that are the actual ways and actions of handling the 

data, have been presented in previous research, and the researchers should choose the 

analysis method that best suits the aim and nature of the research in question (Gunther 

& Hasanen, 2021). When discussing qualitative analysis methods, content analysis is usu-

ally brought up as a method for conducting analysis on qualitative data (Vuori, 2021; 

Gunther & Hasanen, 2021). Content analysis focuses on creating a description of the 

research subject with the examination of the data set (Vuori, 2021). Different forms of 

content analysis, such as coding, thematization and typologies, are traditionally linked 

to qualitative analysis (Gunther & Hasanen, 2021). They can be used as the actual 

method for analysis or as a basis for more specific analysis methods that are based on 

varying theoretical-methodological constructs (Gunther & Hasanen, 2021).  

 

This research utilizes thematic analysis created by Braun and Clarke (2006) and the Gioia 

method presented by Gioia et al. (2012) as the analysis methods. This selection of anal-

ysis methods provided a basis for conducting a data-driven analysis of qualitative data 

that is compatible with the data gathering method of this research (Gioia et al., 2012; 

Braun & Clarke, 2006). The focus of thematic analysis is on identifying commonalities in 

the data, that are the common ways in which people discuss a certain topic. Those com-

monalities form patterns of meaning, that are the themes generated from the data 

(Braun & Clarke, 2012). Part of the analysis is the exploration, analysis and interpretation 

of the patterns (Braun & Clarke, 2006, p. 79).  

 

The Gioia method is a mode of thematic analysis. It is applicable for this type of research 

as it was created specifically for exploring organizations and organizational life (Gioia et 

al., 2012). The focus of the Gioia method is on concept development (Gioia et al., 2012). 

Gioia et al. (2012, p. 16) define a concept as “a more general, less well-specified notion 

capturing qualities that describe or explain a phenomenon of theoretical interest”. In 

addition to concept development, the Gioia method can be utilized to generate new the-

ories, or the developed concepts can be further formulated into measurable constructs 



34 

(Gioia et al., 2012). The main assumptions that the method is based on include the as-

sumption of the organizational world being socially construed and that the members of 

these organizations are capable of comprehending and explaining their actions, objec-

tives, thoughts and experiences (Gioia et al., 2012). The Gioia method requires thorough, 

systematic and analytical work from the researcher, that produces credible research find-

ings (Gioia et al., 2012).  

 

Gioia et al. (2012) state that the analysis is embedded with data gathering, as the first 

stage of analysis includes the exploration and identification of informant terms adhered 

from the data set. This stage of analysis is referred to as 1st order analysis, that aims to 

generate 1st order categories and to label those categories while preserving the inform-

ant terms used (Gioia et al., 2012, p. 20). After the initial 1st order analysis, the genera-

tion of 2nd order themes follows as the 1st order categories are formulated into larger 

entities (Gioia et al., 2012, p. 20). Connections and patterns between the categories are 

explored to produce these 2nd order themes, and their meanings are examined and in-

terpreted (Gioia et al., 2012, p. 20). According to Gioia et al. (2012) a central part of 2nd 

order analysis is to explore the generated themes and whether they propose a possible 

concept regarding the research question. The 2nd order themes are then further refined 

into aggregate dimensions (Gioia et al., 2012).  

 

Following the 1st and 2nd order analysis the formulation of a data structure is carried 

out (see Figure 1.) (Gioia et al., 2012). The data structure is used as a tool to visualize the 

data and to provide structure for further analysis (Gioia et al., 2012). When presented 

with the results of the research it also provides a visual representation of the data and a 

re-enactment of how the analysis proceeded, enhancing the credibility of the analysis 

(Gioia et al., 2012). The final stages of analysis include contrasting the themes and con-

cepts with each other and the data, inspecting whether the formulated data structure 

provides a comprehensive view of the data and contrasting previous research and liter-

ature with the findings (Gioia et al., 2012, p. 21). Revisiting the data and revising the 



35 

themes are allowed, as the method enables this kind of flexibility to the research (Gioia 

et al., 2012). 

 

Acknowledging the influence of the researcher is important when utilizing the Gioia 

method, as their interpretations of the data are central when forming the categories and 

themes and reporting the findings of the research (Gioia et al., 2012). The credibility of 

the research findings is construed by systematic work, providing of the data structure as 

part of the research findings (see Figure 1.) and supported by the extracts drawn from 

the data set (Gioia et al., 2012; Braun & Clarke, 2012). 

 

In this research, the initial steps of analysis after data gathering included the transcrip-

tion of the data set from the interview recordings into written text. This was done so that 

the data set was easier to process in the following stages of the analysis (Gunther & 

Hasanen, 2021). Increasing the quality and credibility of the data set and following the 

guidelines of the Gioia method, the transcription was carried out with precision, so that 

the informant terms and other contents were preserved and remained visible in the data 

in their original form and nature (Gioia et al., 2012; Hirsjärvi & Hurme, 2022). Accordingly, 

familiarization with the data was continued by listening to the recordings, reading 

through the transcriptions and making notes about the data. Reducing the data from 

inessential information that did not hold any value regarding the aim of the research was 

also carried out (Braun & Clarke, 2006, p. 87). 

 

After the initial steps of the analysis, the 1st order analysis was carried out. A large num-

ber of first order categories emerged as a result of manual processing of the data. The 

data set was processed manually throughout the analysis, as it allowed the close explo-

ration and familiarization with the data in accordance with the Gioia methodology (Gioia 

et al., 2012). Following this the 2nd order analysis began with exploring and identifying 

similarities and differences between the categories and formulating the 2nd order 

themes based on these factors. The 2nd order themes were then contrasted with the 

research question, and evaluation of whether they provided relevant information or 



36 

insight into the topic of the research was conducted. This evaluation was done so that 

the essential 2nd order themes could be identified from the data and further successfully 

formed into aggregate dimensions, leaving out themes that did not provide relevant in-

formation or contribute to the purpose of research (Gioia et al., 2012).  

 

After the 1st and 2nd order analysis were conducted and the aggregate dimensions, or 

in this case the attitudinal concepts were formed, a thematic data structure was formu-

lated to visualize the analysis and its findings (see Figure 1.). The thematic structure aims 

to explain what the concepts were based on and how they were formed from the data 

(Gioia et al., 2012). The final stage of the analysis included the reporting of the findings 

in written form (Braun & Clarke, 2012). The findings discuss each attitudinal concept in 

detail, describing the 1st order categories and 2nd order themes significance and contri-

bution to the interpretations made by the researcher. Data extracts are included in the 

findings to create a structure for the analysis and work together to support the storyline 

created in the analysis (Braun & Clarke, 2012). 



37 

4 Findings 

The research findings provide an outlook of the different and varying attitudes the em-

ployees expressed towards the introduction of AI in their organization. The attitudes 

were formed based on the exploration and interpretation of the perceptions, anticipa-

tions and expectations and their meanings, described and expressed by the employees 

regarding AI adoption in the organization. By identifying the similarities and differences 

of these factors, four main attitudinal concepts were formed utilizing the Gioia method-

ology (Gioia et al., 2012). The attitudinal concepts gathered from the data include em-

ployee openness, commitment, cynicism and skepticism towards AI adoption.  

 

The attitudinal concepts and their defining characteristics are covered in the thematic 

structure below (see Figure 1.), including the initial 1st order categories and 2nd order 

themes. The thematic structure provides an overview of the research findings with the 

aim to clarify the defining and differing characteristics of the attitudinal concepts. The 

attitudinal concepts are discussed in the following chapters in more detail. 



38 

 

 

Figure 1. Thematic structure of employee attitudes in AI adoption 

 

 

4.1 Employee openness towards AI adoption 

Employee openness towards AI adoption appeared in the data as employees’ desire for 

changes and development at work as well as eagerness to adopt AI and gain the 



39 

anticipated benefits of AI adoption. It was characterized by positivity and optimism, as 

the employees described their expectations and perceptions regarding AI adoption. Em-

ployee openness towards AI adoption was concluded to portray a more acceptive atti-

tude towards the technology and the changes brought by it. Employees worded and de-

scribed the adoption of AI positively, indicating that the change was welcome, wanted 

and possibly provided benefits and opportunities for them, their work and the organiza-

tion.  

 

(1) I believe it is a really good opportunity, more in a way like, it is in a 
positive way fun to see, where it takes us. And yes definitely, we 
have to keep up with it. We can’t think that we can forever keep 
doing things a certain way like we have before.  (Participant 2) 

 

In extract 1 Participant 2 describes viewing the change as an opportunity and indicates 

a positive, welcoming outlook towards the change. They imply that changes are neces-

sary, showing an understanding of changes being a natural occurrence in organizational 

life. Other employees expressed a different source of openness as they wished for tools 

or other aids that would lighten their workload or release time for them to focus on more 

important tasks, such as serving customers or tasks demanding precision. Some believed 

that changes in general brought positive additions and variety to their work, making their 

work more interesting and enjoyable. This was concluded to indicate employees’ desire 

for changes and development at work in general, that was connected to employee open-

ness towards AI adoption. 

 

Alongside a more acceptive outlook towards changes in general, part of the employees 

expressed curiosity towards AI and its adoption in the organization. Some of them de-

scribed having a personal interest in the topic, and they were already familiar with the 

phenomena and had started experimenting with AI outside of work. The employees 

were keen to get more information about AI and what possibilities it could bring and 

how it would impact their work. They described having an interest to see how AI would 



40 

be implemented into their work and the organization, indicating excitement and eager-

ness towards it.  

 

(2) And like kind of curiosity and interest in the topic. So, I believe I 
would be excited to start using it for the reason as I am interested 
in the topic in general. It would be exciting to see how it would be 
implemented to my own work here. (Participant 7) 

 

In extract 2 Participant 7 describes having a personal interest in the topic and excitement 

to start using AI as part of their work. They express eagerness to see how AI adoption 

will plan out in the organization and how it will impact their work. This was concluded 

to indicate a more open and acceptive attitude towards AI adoption. 

 

Eagerness to adopt AI was also connected to beliefs of employee consideration in the 

planning and execution of AI adoption and the anticipation of AI adoption requiring min-

imal effort from employees. The employees had positive anticipations of the change con-

sequences, as they assumed that reason AI was brought into the organization was to 

benefit them or their work and they felt that they could trust that the organization could 

facilitate a successful adoption of AI that considered the employees’ benefit and inter-

ests. The employees were eager to gain benefits resulting from AI adoption, as they an-

ticipated that AI adoption could result in varying benefits, including augmentation of 

employees in service tasks through AIs knowledge sharing and gathering capabilities, 

enhanced and more effective performance through better system quality and operations, 

and better time management and resource allocation. As the participants believed that 

AI was brought into the organization to provide employees with these possible benefits 

and they would come across minimal challenges while doing so, they were open towards 

its introduction and acceptive of it as part of their work.  

 

(3) Surely, I believe it will only bring benefits. I see it as a positive, that 
they are trying to bring AI into our work too and we get – I mean it 
is only going to be easier if you don’t have to remember everything 
and AI is there to help you. And I think even with that you are 



41 

bringing some AI system into an organization of this size, that is pre-
sent in every store for example, then I trust that by that point it will 
be reviewed and tested enough so that it surely won’t cause any 
harm for us. (Participant 6) 

 

In extract 3 Participant 6 describes their perception of AI as part of their work. They 

describe the adoption of AI as a positive change and anticipate gaining benefits resulting 

from AI adoption. They also believe that AI adoption isn’t likely going to cause any harm 

for them or their work indicating trust in the organization’s capabilities to adopt AI, an-

ticipating that AI adoption would not have a negative impact on them. This was identi-

fied to indicate a more open and acceptive attitude towards AI adoption.  

 

 

4.2 Employee commitment towards AI adoption 

Employee commitment towards AI adoption was conceptualized to form from the em-

ployees felt incentives to support AI adoption and felt responsibilities regarding AI adop-

tion in the organization. It manifested in the data as employees’ urge to act in favor of 

the change and willingness to overcome the changes brought by AI. This meant the em-

ployees sensed incentives to support and actively participate in the change process and 

alter their behaviors accordingly to match the change.  

 

Employees felt incentives to support AI adoption as they described perceiving AI adop-

tion as necessary for organizational survival and success. Part of the employees believed 

strongly that AI adoption was a way for the organization to develop their competitive-

ness and exist in the future. This created an urge for the adoption of AI and made the 

change appear reasonable and necessary from the employees’ perspective. Others had 

a strong belief that AI was “present day” and it inevitably be a part of their work and the 

organization and justified the change based on this assumption. These factors were con-

cluded to bind the employees to the change, as they showed willingness to support it.  

 



42 

(4) But sure, I do also understand it that if AI is really changing the world 
as much as it is claimed to change it then we can’t afford to be late 
with it. We must explore these options so that we don’t notice in 5 
or in 10 years or in 15 years that “Oh damn, this should have been 
done earlier”. So absolutely it is good that we are involved in devel-
opment, exploring, as we should be. So, when someone discovers 
how commerce is revolutionized by AI we are prepared for it. (Par-
ticipant 7) 

 

In extract 4 Participant 7 expresses an understanding of AIs possible transformational 

impact, and they are capable of viewing AI adoption from an organizational perspective, 

as something necessary for organizational survival and success in the future. This created 

an urge for change in the employee’s mind, and as they can justify the change and be-

lieve that it is necessary, they imply support for it. 

 

Other employees were identified to experience commitment towards AI adoption as 

they described experiencing obligations and commitment to their work and the organi-

zation in general. Others experienced incentives to support and act in favor of the change 

to come from above, from the organizational leaders in the form of orders that the em-

ployees were then expected to obey. Other employees felt a strong need to perform and 

succeed in their work, and to preserve their competence, as they wanted to identify as 

a good worker and be viewed by others as such. This inner incentive tied them and their 

actions to support AI adoption in the organization. 

 

(5) If I am good at something, I have always been good at my work. So 
that I want to be able perform at work and like I want to – I can 
admit it if I can’t do something, but I have always been willing to 
learn. So, I think that that is one of my best skills that I am not 
ashamed to ask. And that I want to know what I am supposed to do, 
and I want to know how to do it. So, if it is somewhat overwhelming, 
then I will just have to carry on even if it takes 100 or 1000 repeti-
tions. But I believe it is professional ability, that I am up to date. 
(Participant 9) 

 



43 

In extract 5 Participant 9 describes a want to succeed at their work and to identify as a 

good worker. This inner need for performing in their work and being viewed as a good 

worker promoted the employee’s willingness develop their personal competence to 

match the change and even overcome challenges related to the change process. This 

was concluded to indicate the employee’s commitment towards AI adoption.  

 

Employee commitment towards AI adoption was identified in the data as the employees 

described their responsibilities regarding AI adoption in the organization. Part of the em-

ployees expressed a voluntary willingness to develop their skills and alter their behavior 

accordingly to match the change. The employees experienced that AI adoption posed a 

need for them to develop their skills and most of the participants pleased to do so. Oth-

ers even expressed readiness to take on an active role in the change. Taking on an active 

role meant being self-driven and actively supporting change through their own actions 

and behaviors as well as helping others overcome the change.  

 

(6) Definitely I will need to develop my own skills, yes. I feel like my 
current skills are enough for me to quickly take up new things and 
like it is easy to accustom to new things. For sure it is a place for 
learning new things. (Participant 2) 

(7) I think it’s probably because I view it positively and I think it is a 
really interesting thing and a good addition that will surely enable a 
lot. Then that helps with, that I can surely with my own attitude in-
fluence the surrounding team’s attitude and capability to take the 
new thing as their own. (Participant 2) 

 

In extract 6 Participant 2 recognizes that the change poses a need for personal develop-

ment, and they express readiness and willingness to do so. In the following extract 7 they 

indicate that they see themselves as having an influence on change through their own 

actions. They imply support for the change and willingness to assist other employees 

during the change process. These factors were concluded to indicate Participant 2’s com-

mitment towards the change. 

 

 



44 

4.3 Employee cynicism towards AI adoption 

Employee cynicism towards AI adoption was conceptualized as a more resistive and neg-

ative outlook towards AI adoption in the organization. Cynical attitudes were based on 

fears, threats, concerns and worries that the employees connected to the adoption of 

AI. It was construed on the anticipated unwanted changes to work resulting from AI 

adoption, the anticipated challenges related to AI adoption and the anticipated threats 

resulting from AI adoption. Cynicism was identified to manifest in the data as employee’s 

hesitance to take part in the change and reluctance to support the change through their 

own actions.  This created an atmosphere that AI was unwanted and unwelcome into 

the organization.  

 

Part of the employees felt cautious about the change and expressed various concerns 

regarding the introduction of AI into their work and how AI would change the content or 

nature of work, their work tasks or roles at work. AI adoption was anticipated by the 

employees to negatively influence their job-satisfaction as it could take over parts of 

work, making work duller and unvaried, producing less experiences of succeeding at 

work and decreasing the experienced meaningfulness of work. Other employees felt 

cautious by the introduction of AI into their work, as they felt that AI could increase their 

competence and performance demands. They felt that if AI would take over basic level 

tasks from humans, human employees would then be expected to possess more expert 

level competence, increasing competence demands. Also, some participants worried 

that they would end up competing with AI, if AI would take part in service tasks. The 

employees then felt that they would need to prove their competence to customers as 

well as having to prove their value as employees to the organization.  

 

(8) If something like this came, then certainly there would be less need 
for employees. Then it would probably create some kind of compe-
tition there. Like how much of a service person you need to be your-
self, what to know and do, and how much of a personality you need 
to be in situations and how full on you need to perform in every 
situation. Maybe it also depends on what the premise is for why AI 



45 

is introduced. Like are we talking about a screen of some sort or is 
it just a, like basically an option or is it introduced as more like “Hey 
now you can do everything with AI, you don’t need to talk to any-
one!”. I mean that, you can launch it very differently - are we launch-
ing it as assistance or as replacement. (Participant 6) 

 

In extract 8 Participant 6 discusses AI that is utilized in customer service. They express 

their worry of AI increasing performance and competence demands for employees, if AI 

was to be utilized in service tasks. They feel like they might end up competing with AI or 

having to intensify their performance at work. In the same extract, Participant 6 also 

brings up the relevance of AIs application for their reactions to the change. They implied 

having differing attitudes towards the different applications of AI as they sensed that AI 

that was meant to replace human workers created a threat for employees and therefore, 

they indicated having a more negative attitude towards it.  

 

Relating to the unwanted changes that AI was anticipated to bring to work, part of the 

employees expected that the AI systems or applications would be of low-quality and 

therefore provide poor user-experiences. AI adoption was then feared by the employees 

to result in increased technical challenges, heavier workload and added stress to work. 

Other concerns related to AI technology usage at work included the concern of how AI 

adoption would affect employee roles and responsibilities at work. The employees de-

scribed not wanting to lose control over their work to AI and felt the need for autonomy 

over their work. Part of the employees also expressed troubles trusting AI, as they did 

not perceive AI as adequate to independently perform tasks in the workplace, and de-

scribed having a need to govern AI.  

 

(9) Well I don’t want it to gain too much power. I do enjoy holding the 
strings in my own hands and like for example I like these stores 
where we do a lot of independent work in the sense that I am then 
in control all the time and I know what is going on. Then kind of that, 
if AI was to take too much of that control away from me. (Participant 
12) 

What do you mean by taking control away from you? (Interviewer) 



46 

Like probably that like it would independently do all the background 
things, like everything related to cargo and stuff. Like I would have 
to always go and check them every day like is it correct and does the 
product quantities match and so on. (Participant 12) 

 
In extract 9 Participant 12 expresses the concern of AI possibly decreasing their auton-

omy at work. They describe wanting to be in charge of their work and having a sense of 

autonomy and control. In addition, they indicate distrust towards AI systems perfor-

mance, as they would need to review that the systems have operated correctly. This was 

concluded to signal a negative affinity towards AI and further indicate employee cynicism 

towards AI adoption in the organization. 

 

Other employees anticipated coming across challenges during the AI adoption process, 

as well as when using the systems as part of their work. The employees expressed con-

cerns about their own abilities to accommodate the change, perceiving the adoption 

process as possibly more difficult and uneasy. In addition, previous technical changes in 

the organization were experienced to be challenging. The employees described that the 

previously introduced technologies had not been fully developed to perform adequately 

and therefore the initial stages of the adoption process were experienced as difficult and 

frustrating. In addition, the employees described not having received enough support to 

overcome the changes. This made the employees more reluctant to adopt AI.  

 

(10) Maybe in a way that can AI be like a runaway horse, that like kind of 
gets out of hand. So that am I able to control it and keep the bridles 
with me like I want to and understand to. Because in a way my own 
power of comprehension and understanding of it or AI is quite lim-
ited, so that. And then, however much I would try to gather infor-
mation for myself about it, everything changes all the time and like 
goes forward, that like I know I wouldn’t be able to keep up with it.  

And yes, I – I don’t really have that self-initiative to it. So, I’m hoping 
that there would be, like some mandatory courses here, that we 
would need to go through and pass. (Participant 11) 

 



47 

In extract 10 Participant 11 discusses their personal abilities to overcome the change and 

related concerns that they had addressing AI adoption. They anticipate that they would 

have difficulties keeping up with the rapid pace of development and changes relating to 

AI, as they have limited knowledge about AI and perceive themselves as slower adapters. 

They describe not having a self-initiative to act and support the change, but rather ex-

pect it to come through obligation and mandatory actions ordered from organizational 

leaders. This kind of negative anticipation of AI adoption was connected to employee 

cynicism towards AI adoption. 

 

The third source of employee cynicism towards AI adoption was identified to include the 

possible threats that the employees connected to AI adoption. The concerns expressed 

by the employees included the worry about the change being made on the employees’ 

expense or without careful consideration of employees in the change, indicating distrust 

towards the decision-makers and leaders of the organization. Some of the participants 

expressed concerns about their position in the organization, as they anticipated that AI 

adoption would affect how valuable human work would be.  In relation, fears of AI re-

placing human work were expressed. Part of the employees believed that AI was 

adopted with the pursuit of financial goals in the form of cost reductions. The adoption 

of AI was then concluded to result in replacements of human employees or reduction of 

work hours by the participants. 

 

(11) I’m not sure if it could take over cargo or something like that. But 
then again like will it make us useless here. Like it is so hard to think 
of it like, as I already feel like we do have these tools here already. 
Like if AI was to take over cargo then I would have more time to 
serve customers, like better. But then it would most certainly end 
up with us (employees) being decreased from here if we end up to 
it. Like of course that if it didn’t – I wouldn’t have to be afraid that 
will my work be decreased. (Participant 14) 

 

When discussing the possible use cases of AI in the organization with Participant 14, they 

seem to evaluate the possible benefits and risks of AI adoption. They see that AI adop-

tion would benefit their work by freeing time for them to focus on more important tasks 



48 

but seem to connect the fear of job-displacement to it simultaneously. This fear of job-

displacement seems to override the perceived benefits of AI adoption in the employees’ 

mind and therefore makes them perceive the adoption of AI in a more negative light, 

indicating cynicism towards AI adoption. 

 

 

4.4 Employee skepticism towards AI adoption 

Employee skepticism towards AI adoption was concluded to comprise of the negative 

employee expectations of the success of AI adoption in their organization. The employ-

ees expressed uncertainties and cautiousness, that were concluded to signal distrust in 

the organization’s capabilities to accommodate AI and an anticipation of minimal gained 

benefits resulting from AI adoption. The employees reflected on past experiences in the 

organization and made evaluations of the decision-makers’ and organization’s capabili-

ties based on them. Skepticism characterized a more negative and critical attitude to-

wards AI adoption, that comprised of employee doubts and reservations. 

 

Many participants felt cautious towards the introduction of AI into their work and the 

organization. Their previous experiences with technology adoption in the organization 

were negative, weakening the employees’ trust towards the organization’s capability to 

carry out a successful AI adoption process. The employees made judgements about 

whether the changes were necessary, well designed and well executed, relating to the 

perceived ease of learning and adaptation of previous technical changes in the organiza-

tion. Previous experiences highlighted the organization’s inability to accommodate such 

changes successfully from the employees’ perspective and had left the employees with 

uncertainties regarding future change processes.  

 

(12) I’m excited that it’s explored, I’m excited about it and with a positive 
mindset that we are being brave and exploring it. But maybe relat-
ing to what these system deliveries have been before, that they 
have always been everything else but good and easy. Especially like 
that they may have been, that they are later developed to be good, 



49 

b