ISSN 2463-9281
Izzivi
prihodnosti
Challenges of the Future
Letnik 10, številka 4, november 2025
Volume 10, Issue 4, November 2025
ISSN 2232-5204
155 UPSKILLING OLDER EMPLOYEES IN THE
ARTIFICIAL INTELLIGENCE ERA
Tinkara Žabar,  Aleksander Janeš 
173 IOBSTACLES TO THE IMPLEMENTATION OF
INNOVATIONS IN START-UP COMPANIES
Nerman Ljevo,  Sabina Šehić -  Kršlak
188 AI IS NOT A TOOL: THE IMPACT OF
GROWING AI AGENCY ON THE FUTURE OF
WORK
Alexander van Biezen
KAZALO VSEBINE
TABLE OF  CONTENTS
DODATEK
APPENDIX
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
DOI: 10.37886/ip.2025.007 
 
 
Upskilling Older Employees in the Artificial 
Intelligence Era 
 
Tinkara Žabar1   , Aleksander Janeš 2  
1 University of Primorska, Faculty of Management, tinkara.kodelja@fm.upr.si  
2 University of Primorska, Faculty of Management, aleksander.janes@fm.upr.si 
 
 
Abstract 
Research Question (RQ): What is the effect of new technologies, with an emphasis on artificial 
intelligence (AI), on the need to upskill older employees (50+ years). 
Purpose: The purpose of the research was to carry out a systematic literature review of existing 
research in the field of the effect of AI on the upskilling needs of older employees. 
Method: We performed a systematic literature review across six academic search engines: ProQuest, 
Emerald, Sage Journals, Springer, Research Gate, and Google Scholar. 
Results: Artificial intelligence is significantly transforming the labor market, as it requires constant 
adaptation to new skills and knowledge. AI has a significant effect on older employees, who are 
exposed to greater challenges due to a possible lack of digital skills and sensitivity to change. In this 
context, training and further education are key mechanisms to ensure that skills match the 
requirements of the work environment and the labor market. Organizations must quickly adapt to 
changing requirements by creating a culture of lifelong learning that encourages seniors and other 
employees to improve. Training programs must be based on the specific needs and challenges faced 
by older employees.  
Organization: The research emphasizes the importance of training older employees in the age of AI 
and encourages organizations to create a culture of lifelong learning as part of the organization's 
strategic directions and goals. 
Society: The importance of research for society is reflected in the insight into the involvement of all 
age groups in the possibility of improving knowledge, skills, and attitudes towards the use of modern 
technologies. Organizations and society itself bear the social responsibility to enable older employees 
to successfully integrate into the work environment in the AI era.  
Originality: The research addresses the need to improve the skills of a specific age group in the age of 
AI, where it simultaneously highlights the importance of fostering a culture of lifelong learning in a 
rapidly changing world. The research findings provide guidelines for policymaking in the field of 
training on the national level in the context of an aging workforce and new technologies. 
Limitations/further research: The literature review was limited to six publicly available databases. In 
the article, older employees were considered as people in the labor process older than 50 years. We 
must emphasize that older employees differ from each other in terms of education, economic, social, 
and other circumstances. Further research should investigate the effect of new technologies 
regarding the specific circumstances mentioned in this age group. 
 
Keywords: knowledge society, upskilling, knowledge management, retraining, older employees, 
artificial intelligence, lifelong learning. 
 
 
Received: 2025-06-25, revised: 2025-07-02, accepted: 2025-10-09  155 
as Original Review Paper 
 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
1 Introduction 
Technological advances, particularly artificial intelligence (AI), are profoundly reshaping job 
roles and work structures. Organizations are integrating AI technology into their business to stay 
competitive. However, the introduction of such technologies into business processes 
contributes to the exposure to inequalities such as digital skills, employment volatility, the 
impact of automation and robotics, in parallel with processes of job destruction and creation 
(Cramarenco et al., 2023, pp. 732). These transformations underscore a pressing imperative for 
large-scale workforce upskilling, with projections indicating that by 2050, approximately 50% of 
all employees will require upskilling as a result of the use of emerging technologies in the 
workplace. To ensure meaningful employee engagement, organizations must provide learning 
opportunities that equip workers for these transitions. Such efforts contribute to the 
development of an inclusive and resilient knowledge-based economy in which all individuals can 
participate meaningfully (Li et al., 2023, pp. 1697).  
New technologies often have a stronger impact on older workers, as they tend to have less 
developed digital competencies needed to manage AI in comparison to their younger colleagues. 
In the literature, older employees are generally defined as individuals who are active in the labor 
market. The age at which someone is defined as an older employee is not universally agreed 
upon. Most studies refer to those aged 50 and above (Novak, 2023, pp. 23), with definitions 
typically starting between age 45 and 55, and with the upper age limit often left unspecified 
(Krašovec, 2015, pp. 30). In our work, we refer to older employees as those who are over 50 years 
of age and are actively participating in the labor force.  
Despite the valuable professional experience that older workers possess, many of them lack 
digital skills necessary to effectively engage with AI systems, making targeted retraining and 
upskilling efforts essential (Tiku, 2023; Chetty, 2023). A common misconception is that older 
people are disinterested in training. However, participation many times correlates with the right 
training format. When learning formats are adapted to meet older employees’ preferences, they 
are significantly more inclined to engage in further education (Zwick, 2015, pp. 146). Many 
researchers point out that creating a culture of lifelong learning in the workplace is crucial for 
fostering skills development in a rapidly changing world and labor market demands (Tiku, 2023; 
Li et al., 2023; Vuorenkoski et al., 2018; Pradhan et al., 2023).  
Based on a systematic literature review, we aimed to answer the following research question: 
What is the effect of AI on the need to upskill older workers? In the fourth section, the main results 
will be outlined based on the articles included in the literature review. 
 
 156 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
2 Theoretical framework 
Digital technologies, particularly AI, commonly defined as computer programs capable of 
performing tasks that typically require human reasoning, represent a key advance in 
technological development (Bruun and Duka, 2018, pp. 1). With these characteristics, AI plays a 
transformative role and significantly reshapes the nature of work by influencing job structures, 
business operations, and employee monitoring (Classen et al., 2018, pp. 23). One of the key 
contributions of AI is that it enables the automation of certain processes, such as repetitive, 
everyday tasks that do not require expert knowledge. This, in turn, allows employees to focus on 
tasks that require human attention and understanding (Haizir, 2022, pp. 7). Parallel to 
technological advances in the workplace, the workforce is also declining. Therefore, the adoption 
of new technologies is more significant in countries with rapidly aging populations, where 
innovation is often driven by the need to compensate for a shrinking workforce (Acemoglu and 
Restrepo, 2021).  
As digital technologies continue to evolve and become more integrated into the workplace, active 
participation in the labor market increasingly depends on the possession of advanced digital 
competencies (Komp-Leukkunen et al., 2022, pp. 37–38). The integration of AI into companies 
also introduces challenges, such as creating new tasks and threatening job losses for employees 
with limited digital skills, which causes a change in the needs for skills or competencies in the 
labor market (OECD, 2024). As a result of the integration of AI into work processes, digital literacy 
has become one of the most desirable characteristics that an employee can have (Bokek-Cohen, 
2018, pp. 21).  
Integration of new technologies also presents a big shift in the labor market. Bruun and Duka 
(2018, pp. 3) have researched the impact of the level of automation on the labor market and how 
the economy will adapt to it by 2038, where they used cheese terminology. Based on this, they 
have developed three scenarios:  
- Stalemate: The AI revolution will be much smaller than expected and will not change the 
nature of work. The economy does not need to adapt; employment in 2038 will look the 
same as today. In this scenario, there is no cause for concern, and governments can 
continue business as usual. 
- Check: Despite the wave of automation, the economy can move and adapt, allowing new 
jobs to be created to replace those lost. The transition may cause some initial chaos in 
the labor market, but after a period of discomfort, stability will follow. 
- Checkmate: The AI revolution will lead to rapid job losses as the economy, governments, 
and individuals will fail to keep up. The main concern is that the economy might not adapt 
quickly enough. Reflected in the increased risk of technological unemployment and 
social instability. 
 
 
 157 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
It is difficult to determine which of these scenarios will occur in the future. According to Bruun 
and Duka (2018, pp. 5), it is best to prepare for the worst-case scenario, namely the checkmate 
scenario. One of the finest strategies to prepare for unknown scenarios is investing in upskilling 
and skills development programs to keep people competitive, as AI shapes the functioning of the 
labor market. These programs would enable the current working-age generation, who are most 
affected by AI, to acquire the relevant skills to be more confident in operating in the labor market. 
It is assumed that future generations will need to be retrained several times in their working lives 
in order to keep pace with technological progress. Learning will therefore not end after tertiary 
education but will continue throughout an individual's life course (Bruun and Duka, 2018, pp.10). 
Digital technologies are not only transforming the economy and sectors but also creating new 
jobs and demanding new skills. As a result, employers are seeking individuals who possess 
technical and digital skills, as well as skills that automation has not yet been able to replicate, 
such as cognitive and social skills (Lincoln, 2017, pp. 7). Taking the uncertainty of the impact of 
AI on jobs into account, the most effective response lies in developing systems that promote both 
individual capabilities and societal learning. A good level of basic skills and a broad knowledge 
base will thus be reflected in citizens' learning opportunities. However, in a rapidly changing 
economy, it is necessary to understand that skills that currently ensure high wages may soon lose 
their relevance (Vuorenkoski et al., 2018, pp. 39–40). The present workforce is not yet ready to 
embrace new technology due to a lack of relevant skills. Consequently, organizations must invest 
in upskilling and encourage workers to incorporate AI into their everyday tasks (Pradhan and 
Saxena, 2023, pp. 181). 
New technologies impact age groups in distinct ways. Younger individuals generally possess 
stronger capabilities for understanding and adapting to technological developments compared 
to older age groups (Tiku, 2023, pp. 3). Older workers, particularly low-skilled, are often defined 
as having lower digital competencies. This can lead to a sense of threat from advances in 
automation, creating a skills mismatch and even leading to earlier retirement (Aisa et al., 2023, 
pp. 9). Besides the lack of digital skills, older workers also face age discrimination, which affects 
their employability (Lee et al., 2008). As a response to these intersecting challenges, upskilling 
programs represent a key component of broad workforce strategies. Employees are generally 
interested in training programs, but they do not attend them due to various socio-economic 
barriers such as the cost of education or family commitments (Bianco, 2021, pp. 9).  Older 
population in the workplace has greater difficulties in transitioning to new job roles due to 
automation, while the adult learning participation is generally lower, making it harder to pursue 
upskilling and re-training for this age group (Nedelkoska and Quintini, 2018). There is also a lack 
of incentives for employers to invest in the training of older employees, affected by low 
participation rates (Alcover et al., 2021). 
 
 158 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
It should be emphasized that the workplace has a significant impact on older workers in the 
process of digitization, through the implementation of information and communication 
technologies (ICT), the possibility to work remotely, the training offered, and the attitudes of 
employers and managers towards employees (Komp-Leukkunen et al., 2021, pp. 49). It is crucial 
to ensure an inclusive work environment for older employees by allowing them to develop skills, 
career development, flexible education, training programs (Cramarenco et al., 2023, pp. 747), 
and intergenerational cooperation (Waligóra, 2024, pp. 103). However, the lack of basic skills 
remains a key barrier to the uptake of AI and poses a challenge for adults with limited digital skills. 
It reflects the need for tailored skills policies, as this age group, together with low-skilled workers 
and people living in rural areas, face the lowest participation rates in training programs (OECD, 
2024). Upskilling older employees enhances their competitiveness in the labor market (Trunkina 
et al., 2019, pp. 522). At the same time, organizations also benefit from investing in their 
employees. A failure to support older employees in adapting to new technologies risks the loss 
of valuable knowledge, experience, and perspective that this demographic brings to the 
workplace (Zwick, 2015, pp. 146).  
Adult education trainers must take into account the specific motivations of older employees that 
affect their participation in such programs. While this age group wants flexible training programs 
that provide practical and immediately relevant knowledge (Zwick, 2015, pp. 146). In the case of 
specific training for the use of AI, this training should include specific knowledge that can be 
applied in the workplace of the current employer, while developing critical thinking and meta-
cognitive skills (Chetty, 2023, pp. 9). 
Given the fast-paced evolution of skill requirements in the labor market due to technological 
advancement, fostering a culture of lifelong learning within organizations should be considered 
a strategic priority and integral to their long-term objectives (Lincoln, 2017; Bianco, 2021; Li et 
al., 2023). Embedding continuous learning into organizational culture supports workforce 
adaptability and enhances resilience in rapidly changing environments.  
3 Method 
The research is based on a systematic literature review, which allows a structured and 
comprehensive analysis of existing research on the chosen topic. The scientific research articles 
were retrieved from the following databases: ProQuest, Emerald, Sage Journals, Springer, 
Research Gate, and Google Scholar. The keywords used to search the databases were "older 
employees" OR "older workers" AND "upskilling" OR "skills" AND "artificial intelligence" OR "AI" 
OR "artificial intelligence era". The data collection took place between November 2024 and 
January 2025. The total number of articles found by keyword was 15.888. 
In our literature review, we set inclusion criteria that guided our research collection. For the 
inclusion criteria, we considered that: 
 
 159 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
- The article was published between 2014 and 2024, 
- The text was published in English,  
- The availability of the full text; and  
- The article focused on the topic of upskilling older workers in relation to new 
technologies, with a focus on AI. 
 
The systematic literature review followed the Preferred Reporting Items for Systematic Reviews 
and Meta-Analyses (PRISMA) principles (Page et al., 2021). The keywords mentioned above were 
entered into the databases, where duplicate records (n=658) were removed from the search. This 
was followed by a screening of titles and abstracts based on the topic under consideration. 
Publications that did not address our research objective were excluded, and we continued 
reading the full texts. 13 publications were included in the final analysis according to the 
inclusion criteria and the research question. The publication collection procedures are shown in 
Figure 1. 
Figure 1 
The process of systematic selection of publications according to PRISMA 
 
 
  
 
 
 
 160 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
4 Results 
Tabel 1 provides a synthesis of the findings from 13 articles that met the inclusion criteria. The first column lists the authors and year of publication, 
followed by columns detailing the (original) title, methodology, research region, publication type, and a summary of key findings.  
Tabel 1 
Author and year of publication, title, methodology, research region, publication type, and key findings 
 
Author and year    
of publication Title Methodology Research region Publication type Key findings 
 
Literature review and Europe  Journal Article  -Older workers may lose their jobs or have their job 
quality reduced due to the interaction between 
Alcover et al., Aging-and-Tech Job concept development 
Vulnerability: A Proposed ageing and job automation. 
2021 Framework -Both the individuals and the organizations are 
responsible for reducing the vulnerability of older 
employees. 
Literature review and Europe Journal Article  -Two simultaneous processes in developed 
secondary data analysis countries - ageing populations and technological 
Ageing Workers and Digital  innovation. 
Bianco, 2021 Future -Higher educated workers are more willing to 
participate in training programs. 
-Lifelong learning to keep older workers competitive 
in the labor market. 
Conceptualizing Literature review - Israel  Journal Article  -Digital skills as a signal to the employer, informing 
Bokek-Cohen, Employees’ Digital Skills as Spence's signaling theory them of the (future) employee's capabilities, 
2018 Signals Delivered to adaptability, and potential within the organization.  
-Older employees need to compensate for their 
Employers proportionally higher age by learning new skills. 
Quantitative-secondary Europe Journal Article  -AI has a dual impact on the decision to retire early, 
Casas and The Impact of Artificial data from the European depending on the education of workers and the 
Román, 2024 Intelligence in the Early Survey on Health, Ageing nature of the occupation.  
Retirement Decision and Retirement -More educated workers are more able to adapt to 
technologies. 
 
 161 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
Literature review South Africa Journal Article  -The development of the digital economy has 
encouraged companies to train employees in AI 
literacy. 
AI Literacy for an Ageing -Older employees are at a disadvantage compared 
to younger employees who have more advanced 
Chetty, 2023 Workforce: Leveraging the 
Experience of Older knowledge of the latest technologies.  
-AI can enable older workers to participate 
Workers strategically in the digital economy.  
-Training programs for working with AI for older 
employees should consider that they contain 
specific knowledge for their job.  
The Impact of AI on Literature review Europe  Journal Article -Technological advancement is affecting jobs, 
leading to (digital) skills gaps and job volatility. 
Cramarenco et Employees’ Skills and 
Well-being in Global Labor -Key solutions include continuous skills upgrading, 
al., 2023 Markets: A Systematic financial support for lifelong learning, tax incentives 
for employers, grants for flexible training programs, 
Review and international partnerships to promote mobility. 
Is the Obsolescence of the Literature review Europe Book Chapter -Digital technology is reforming the organization of 
Skills of Older Employees workplaces and requires employees to develop new 
Cros et al., 2021 an Inevitable skills.  
Consequence of -This development is not given to a vulnerable 
Digitalization? group. 
Literature review and America  Book Chapter -The "greying" of workplaces is forcing a paradigm 
panelist interviews shift in workplace policies.  
-The ageing workforce is forcing organizations to 
A Hiring Paradigm Shift better understand the acceptance and retraining of 
Hughes et al., through the Use of an older workforce. 
2019 Technology in the -Strategies to meaningfully engage an ageing 
Workplace workforce: learning and development, knowledge 
transfer, and career path development. 
-Lifelong learning is becoming a growing trend 
across all ages. 
Quantitative data from China  Journal Article  
Does Artificial Intelligence -The impact of AI on workplace learning affects the 
the China General Social 
Promote or Inhibit On-the- older workers, women, and those with lower levels 
Li et al., 2023 Survey (CGSS) of education, as well as those without employment 
Job Learning? Human  
Reactions to AI at Work contracts and with less job autonomy and work 
experience.  
 
 162 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 -Based on this, companies and governments should 
pay more attention to these employees and 
 consider providing them with more opportunities 
for training and job protection. 
Quantitative and Europe Report -Elderly people are exposed to stereotypes and 
qualitative discrimination.  
An Ageing Workforce in The 
Lincoln, 2017 Digital Era: Older Workers, -Employers should provide more training and 
Technology and Skills development for older people, create a lifelong 
culture, and tailor training programs for the older 
age group. 
Literature review Europe Journal Article  -The age-related digital divide affects how 
The Impact of Artificial individuals perform their jobs and how they upskill 
Morandini et al., Intelligence on Workers' and reskill.  
2023 Skills: Upskilling and -Organizations that do not invest in upskilling older 
Reskilling in Organisations employees may lose the necessary skills, 
experience, and perspectives. 
Qualitative case study Japan, USA, and India Journal Article -Older employees can acquire and improve their 
digital skills under the right conditions. 
AI-Induced Labor Market -Upskilling and AI can lead to longer working lives 
for older employees.  
Tiku, 2023 Shifts and Aging Workforce -The USA, Japan, and India are trying to bridge the 
Dynamics: A Cross-
National Study digital divide with strategies such as 
intergenerational cooperation, creating inclusive 
workplaces, and creating opportunities for lifelong 
learning. 
Literature review – Russia  Conference Paper -Increasing the competitiveness of older people 
Increasing the secondary analysis of requires a system of training and retraining. 
Trunkina et al., Competitiveness of Older statistical data -Upskilling of the older generation can lead to a 
2019 Age Groups in the paradigm shift in ageism, which creates a 
Digitalization Environment perception of older people as a productive part of 
society. 
 
 163 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
The number of relevant publications analyzing the upskilling of the older workforce concerning 
newer technologies, such as AI, has been increasing over the years. No relevant articles were 
found from 2014 to 2016. An increased number of publications was noted in 2023 that can be 
linked to the launch of one of the most well-known generative AI engine - Chat GPT, in 2022. The 
number of articles by year of publication is shown in Figure 2. 
 
Figure 1  
Number of publications based on the publication date  
5
4
3
publications
2
1
0  
2017 2018 2019 2020 2021 2022 2023 2024
 
The relevance of the topic is reflected in the two simultaneous processes, an ageing population 
and new technological innovations entering the workplace (Alcover et al., 2021; Bianco, 2021; 
Nedelkoska and Quintini, 2018). Older employees often have less developed digital 
competencies, which can hinder their effective use of new technologies (Tiku, 2023, pp. 3). This 
increases the digital divide, which widens the gap between individuals in terms of access and use 
of ICT (Dolničar et al., 2002, pp. 83). Upskilling older workers is crucial to maintain their 
competitive advantage in the labor market (Trunkina et al., 2019, pp. 522) and to reduce the 
chances of early retirement (Casas and Román, 2024) by enhancing their adaptability to 
technological changes. The introduction of new technologies in the workplace encourages 
organizations to create a culture of lifelong learning, which is key to adapting to the new labor 
market (Tiku, 2023; Li et al., 2023; Vuorenkoski et al., 2018; Pradhan et al., 2023).   
5 Discussion 
AI can be understood as a socio-technical construct, shaped through the dynamic between users 
and machines or technology (Orr and Davis, 2020). Despite these findings, researchers, 
academics, and practitioners have also raised concerns about the lack of methods to 
incorporate accountability in socio-technical systems (Verdiesen et al., 2021). To address 
awareness about this gap, there is a need to link the design, production, and implementation 
phases of AI development, including its governance initiatives (Birkstedt et al., 2023, pp. 154). 
These developments are not occurring in isolation, rather, they are profoundly reshaping the 
nature of work as technological advancements are raising the necessity for the acquisition of new 
 
 164 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
and more complex skill sets across the workforce (Lincoln, 2017; Alcover, 2021). In this context, 
the development of digital competencies emerges as a critical precondition for the effective 
adoption and management of technological tools and systems (Cramarenco et al., 2023). 
However, some vulnerable groups, such as older workers, are many times left out of such training 
(Cros et al., 2021). This can even widen the digital divide, while this age group has, on average, 
less developed digital competences compared to their younger counterparts (Tiku, 2023; Janeš 
et al., 2023). Older employees are in the context of digital transformations defined as 'digital 
immigrants' who had to learn how to use digital technologies later in life (Bokek-Cohen, 2018). 
Conversely, younger employees are often characterized as 'digital natives' due to their early 
exposure to technology, which is assumed to result in more advanced digital competencies (Tiku, 
2023). Furthermore, they are expected to possess more up-to-date knowledge of digital tools and 
systems, giving them a relative advantage compared to older employees in the labor market 
(Chetty, 2023).  
Moreover, digital competences serve as a signal of adaptability and employability in the labor 
market for employers (Bokek-Cohen, 2018; Alcover et al., 2021). Komp-Leukkunen et al. (2022) 
have framed digital competencies as a crucial part of general labor market access. It is important 
to recognize that the disadvantaged position of older employees in the labor market cannot be 
attributed solely to lower levels of digital competences (Chetty, 2023; Cros et al., 2021; 
Morandini et al., 2023; Li et al., 2023), but also to challenges in accessing training, due to various 
prejudices, stereotypes, bias, and discrimination related to age (Alcover et al., 2021; Lincoln, 
2017). 
Therefore, new digital technologies in the workplace encourage older employees to upskill to 
remain competitive in the evolving labor market (Alcover, 2021). Acquiring digital competencies 
not only enhances their employability but also serves as a signal to employers, indicating 
capabilities, flexibility, and the potential of the worker for the organization (Bokek-Cohen, 2018). 
Based on that, older employees need to 'compensate' for their age by investing in learning new 
skills that may not have been previously needed. On the other hand, the ageing of the workforce 
is forcing organizations to better understand the needs of this age group and to empower them 
(Hughes et al., 2019). Companies and policymakers need to consider providing more 
experiences to vulnerable groups such as older employees, women, and those with less 
autonomy. However, people who are highly educated are more willing to adapt and train in the 
use of new technologies (Bianco, 2021; Casas and Román, 2018). Training should also be 
tailored to the needs of the older age group (Chetty, 2023), as this is the only way to ensure that 
skills do not become obsolete (Cross, 2021). If older employees are going to upskill, they will be 
able to participate in the digital economy as strategic decision-makers (Chetty, 2023). It is crucial 
to develop suitable systems for training and retraining employees, which include programs for 
the development and implementation of appropriate programs aimed at the active inclusion of 
the older population in the digital economy and knowledge society (Trunkina et al., 2019). Older 
 
 165 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
employees are at a disadvantage in comparison to younger ones looking for a new job (Morandini 
et al., 2023). If some adjustments were taken, this could lead to advanced skills and appropriate 
knowledge, resulting in prolonging the working life of adults (Tiku, 2023). However, this 
relationship appears to be influenced by workers’ educational backgrounds. Casas and Román 
(2018) emphasize that exposure to AI reduces the likelihood of early retirement of more educated 
workers, suggesting that education plays a mediating role in how digital transformation affects 
labor market participation in later life (Casas and Román, 2018). Based on that, there is a need 
for more structured and systematic research on how organizations can effectively adapt to 
technological changes in the workplace and support the continuous upskilling of older 
employees, considering their special needs and backgrounds. Firstly, there should be rigorous 
empirical research about older employees and their attitudes toward AI and AI training, which 
would give practical guidelines for employers on how to design and implement training programs 
effectively for the ageing workforce. Further research should prioritize longitudinal studies that 
explore the participation and experiences of older employees in the workplace and training based 
on suggested guidelines from previous studies. The study mentioned offers a critical insight into 
the long-term effectiveness of age-sensitive interventions.   
As a key strategy for promoting the development of people in the knowledge society and 
competitiveness in the labor market, researchers (Bianco, 2021; Cramarenco et al., 2023; Huges 
et al., 2019) suggest lifelong programs. Meanwhile, Tiku (2023) also highlights the importance of 
intergenerational cooperation and the creation of inclusive workplaces as a strategy for bridging 
the digital divide. It should be emphasized that intergenerational cooperation has two sides of 
the value. Firstly, older employees can gain skills in using modern technologies from younger 
colleagues, which can significantly contribute to supporting work in organizational processes. 
Secondly, younger employees can gain important insight and knowledge about the content of 
work that older employees have developed and mastered over the years. The adoption of such 
approaches and strategies holds the potential to challenge the "age paradox", fostering a shift in 
how older employees are perceived, not as passive or obsolete, but as active and valuable 
contributors to society (Trunkina et al., 2019). Despite the emphasis on the importance of 
adjusted strategies, none of the articles included in the literature review research on the personal 
views of older employees on new technologies. Based on that, we suggest a qualitative study that 
assesses the personal views of older employees about managing AI systems in the workplace. 
The literature review has enabled us to answer our research question: what is the impact of AI on 
the training needs of older employees? New technologies such as AI encourage older employees 
to constantly improve their skills, as upskilling remains one of the most effective strategies for 
maintaining competitiveness in the labor market. Companies are introducing new technologies 
into their operations, which can be understood as pressure on competitiveness, forcing 
companies to update their work processes with modern technologies; if not addressed, this 
could threaten the organization's continued viability. Moreover, with demographic changes, 
 
 166 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
prolonged working lives, and integration of AI in the workplace, training is particularly crucial for 
older employees, as they generally have less developed digital skills to manage new 
technologies. This can also help them to overcome fears, mistrust, and resistance when 
integrating modern technologies into the workplace and serve as a strategy to prolong working 
lives. To sum up, cultivating the culture of lifelong learning is an imperative, as it supports 
continuous skill development among employees of all ages in response to the evolving nature of 
work. Furthermore, training programs should be tailored to the specific needs of participants, 
considering factors such as age, existing digital competencies, and the nature of their work.  
6 Conclusion 
Nowadays, two processes are taking place simultaneously: rapid technological development 
and the aging of the workforce. This article investigates the impact of new technologies, with an 
emphasis on AI, on older employees and the need to upskill them. Generally, older employees 
have less developed digital competencies needed to manage new technologies, which are 
transforming jobs and creating new demands. Without upgrading their knowledge and skills, they 
may become uncompetitive in the labor market, face early retirement, or even job loss. Based on 
that, older employees must adapt to new forms of work and engage in upskilling or retraining 
initiatives.  
Older employees are not a homogeneous age group, as they differ in education, social, 
economic, and other circumstances. Higher-educated older employees tend to be more willing 
to participate in training programs. However, it is essential to emphasize that training programs 
must be tailored to the needs of older employees, as their effective participation depends on 
such adaptation. Moreover, investing in the training of older workers can significantly contribute 
to organizational efficiency and overall success. If organizations do not invest in training older 
employees, they may lose important knowledge and information that older employees possess. 
Additionally, training enables older employees to participate strategically in the digital economy 
and increase their competitiveness within the labor market. Consequently, it is imperative for 
organizations to foster a culture of lifelong learning that extends beyond the older generation due 
to rapid technological progress to equip the workforce with the relevant skills. Furthermore, 
fostering intergenerational cooperation and cultivating inclusive workplaces are also important 
components in the knowledge society. The development of digital skills and knowledge of AI 
management can enable older employees to extend their working lives, increase 
competitiveness in the labor market, and play a strategic role in decision-making. 
A systematic literature review focuses on the impact of new technologies on a specific age group. 
The relatively low number of articles included in the research and the increase in the number of 
publications after 2022 indicate the relevance of the topic and the need for further research in 
this area. It should be emphasized that in the article, we consider older employees as people who 
are over 50 years of age. However, older employees are not a homogeneous group defined solely 
 
 167 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
by chronological age. Building on these considerations, further research should study older 
employees in relation to their educational structure, social, economic, and other factors that 
might influence their capacity to adapt to technological change. In parallel, further studies 
should examine how workplaces and training programs can be more effectively adapted to the 
specific needs of an ageing workforce within rapidly evolving technological environments. This 
would enable researchers to formulate practical guidelines for employers seeking to support 
digital inclusion across all age groups. In addition, qualitative research focusing on the personal 
experiences of older employees when they are managing new technologies would offer valuable 
insights into their coping strategies, perceptions, and barriers faced by a specific age group. 
The article has an important contribution for managers, as it emphasizes the importance of 
creating a culture of lifelong learning to bridge the digital divide among employees and empower 
the workforce with the essential skills in the modern labor market. Older employees bring a 
wealth of experience and knowledge that can greatly benefit both organizations and society. By 
providing targeted training, their competitiveness and value in the labor market can be 
significantly enhanced.  The research also raises the question of how training programs for this 
generation should be designed to effectively support their adaptation to new technologies. 
Furthermore, the findings have broader implications for policymakers, offering guidance for 
shaping national training strategies in response to the dual challenges of workforce ageing and 
technological advancement.  
The research also has some limitations. Firstly, we were limited to articles published only in the 
English language. Secondly, we have defined older employees as a homogenous group, where 
we did not include the possible variables in this age group, which opens up a new research area.  
 
References 
1. Acemoglu, D., & Restrepo, P. (2021). Demographics and automation. The Review of Economic 
Studies, 89(1), 1–44. 
2. Aisa, R., Cabeza, J., & Martin, J. (2023). Automation and aging: The impact on older workers in the 
workforce. The Journal of the Economics of Ageing, 26, 1–12.  
3. Alcover, C.-M., Guglielmi, D., Depolo, M., & Mazzetti, G. (2021). Aging-and-Tech Job Vulnerability: A 
proposed framework on the dual impact of aging and AI, robotics, and automation among older 
workers. Organizational Psychology Review, 11(2), 175–201.   
4. Bianco, A. (2021). Ageing workers and digital future. Rivista trimestrale di scienza 
dell'amministrazione, 3(3), 1–22.  
5. Birkstedt, T., Minkkinen, M., Tandon, A., & Mäntymäki, M. (2023). AI governance: themes, knowledge 
gaps and future agendas. Internet Research, 33(7), 133–167. 
6. Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T., Mulrow, C.,… Moher, D. (2021). 
The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. British Medical 
Journal.   
 
 168 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
7. Bokek-Cohen, Y. (2018). Conceptualizing employees’ digital skills as signals delivered to employers. 
International Journal of Organization Theory & Behavior, 21(1), 17–27.  
8. Bruun, E.P.G., & Duka, A. (2018). Artificial Intelligence, Jobs and the Future of Work: Racing with the 
Machines. Basic Income Studies, 13(2), 1–15. 
9. Casas, P., & Román, C. (2024). The impact of artificial intelligence in the early retirement 
decision. Empirica, 51, 583–618.  
10. Chetty, K. (2023). AI literacy for an ageing workforce: Leveraging the experience of older 
workers. OBM Geriatrics, 7(3), 1–17. 
11. Classen, J., Wegemer, D., Patras, P., Spink, T., & Hollick, M. (2018). Anatomy of a Vulnerable Fitness 
Tracking System: Dissecting the Fitbit Cloud, App, and Firmware. In T. Ploetz & L. Mamykina (Eds.), 
Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 2(1), (pp. 1–
24). ACM digital Library.  
12. Cramarenco, R. E., Burcă-Voicu, M. I., & Dabija, D.-C. (2023). The impact of artificial intelligence (AI) 
on employees’ skills and well-being in global labor markets: A systematic review. Oeconomia 
Copernicana, 14(3), 731–767. 
13. Cros, F., Bobillier Chaumon, M.-E., & Cuvillier, B. (2021). Is the obsolescence of the skills of older 
employees an inevitable consequence of digitalization? In M.-E. Bobillier Chaumon (Ed.), Digital 
transformations in the challenge of activity and work: Understanding and supporting technological 
changes (pp. 169–181). Retrieved 23 March 2024 from 
https://doi.org/10.1002/9781119808343.ch13. 
14. Dolničar, V., Vukčevič, K., Kronegger, L., & Vehovar, V. (2002). Digitalni razkorak v 
Sloveniji. Družboslovne razprave, 18(40), 83–106.   
15. Hughes, C., Robert, L., Frady, K., & Arroyos, A. (2019). A Hiring Paradigm Shift through the Use of 
Technology in the Workplace. In E. Parry (Ed.), Managing Technology and Middle- and Low-skilled 
Employees: Advances for Economic Regeneration, 49–59. Retrieved 19 January 2024 from 
https://doi.org/10.1108/9781789730777. 
16. Janeš, A., Madsen, S. S., Saure, H. I., Lie, M. H., Gjesdal, B., Thorvaldsen, S., ….Klančar, A. (2023). 
Preliminary Results from Norway, Slovenia, Portugal, Turkey, Ukraine, and Jordan: Investigating Pre-
Service Teachers’ Expected Use of Digital Technology When Becoming Teachers. Education 
Sciences, 13(8), 783.  
17. Komp-Leukkunen, K., Poli, A., Hellevik, T., Herlofson, K., Heuer, A., … Motel Klingebiel, A. (2022). 
Older workers in digitalizing workplaces: A systematic literature review. The Journal of Aging and 
Social Change, 12(2), 37–59.  
18. Krašovec, S. J. (2015). Izobraževanje in usposabljanje starejših delavcev–mednarodna 
primerjava. Andragoška spoznanja, 21(2), 29–46. 
19. Lee, C. C., Czaja, S. J., & Sharit, J. (2008). Training older workers for technology-based 
employment. Educational Gerontology, 35(1), 15–31. 
20. Li, C., Zhang, Y., Niu, X., Chen, F., & Zhou, H. (2023). Does artificial intelligence promote or inhibit 
on-the-job learning? Human reactions to AI at work. Systems, 11(114), 1–26. 
21. Lincoln, J. (2017). An Ageing Workforce in The Digital Era: Older Workers, Technology and 
Skills. Business in the Community, A Business in the Community report, supported by Tata 
Consultancy Services. UK: The Prince’s Responsible Business Network, Tata Consultancy Services. 
 
 169 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
Retrieved 30 January 2025 from https://www.bitc.org.uk/wp-content/uploads/2022/12/bitc-report-
age-ageing-worforce-digital-era-march20.pdf. 
22. Morandini, S., Fraboni, F., De Angelis, M., Puzzo, G., Giusino, D., & Pietrantoni, L. (2023). The impact 
of artificial intelligence on workers’ skills: Upskilling and reskilling in organisations. Informing 
Science, 26, 39–68.  
23. Nedelkoska, L., & Quintini, G. (2018). Automation, skills use and training (Vol. 202). Paris: OECD 
Publishing. 
24. Novak, V. (2014). Izzivi dolgožive družbe: staranje prebivalstva, trg dela in ravnanje s starejšimi 
zaposlenimi. In M. Bernik (Ed.), Transformacija kadrovskega managementa (pp. 19–44). Maribor: 
Univerza v Mariboru. 
25. OECD. (2024). Training Supply for the Green and AI Transitions: Equipping Workers with the Right 
Skills, Getting Skills Right. Paris: OECD.  
26. Orr, W., & Davis, J. L. (2020). Attributions of ethical responsibility by Artificial Intelligence 
Practitioners. Information, Communication and Society, 23(5), 719–735. 
27. Pradhan, I.P., & Saxena, P. (2023). Reskilling workforce for the artificial intelligence age: Challenges 
and the way forward. In The Adoption and Effect of Artificial Intelligence on Human Resources 
Management, Part B (pp. 181–197). Emerald Publishing Limited.  
28. Tiku, S. (2023). AI-Induced Labor Market Shifts and Aging Workforce Dynamics: A Cross-National 
Study of Corporate Strategic Responses in Japan, USA, and India. SSRN Electronic Journal. 
29. Trunkina, L. V., Kipervar, E. A., & Mizya, M. S. (2019). Increasing the competitiveness of older age 
groups in the digitalization environment. In International Scientific and Practical Conference on 
Digital Economy (ISCDE 2019) (pp. 236–239). Atlantis Press. 
30. Verdiesen, I., Tubella, A. A., & Dignum, V. (2021). Integrating comprehensive human oversight in 
drone deployment: a conceptual framework applied to the case of military surveillance drones. 
Information (Switzerland), 12(9), 1–13. 
31. Vuorenkoski, V., Lehikoinen, A., Hakola-Uusitalo, T., & Urrila, P. (2018). Learning and skills in 
transition. In O. Koski & K. Husso (Eds.), Work in the age of artificial intelligence: Four perspectives 
on the economy, employment, skills and ethics. Work in the age of artificial intelligence (pp. 37–46). 
Publications of the Ministry of Economic Affairs and Employment.   
32. Waligóra, Ł. (2024). Employees' age diversity - between supportive workplaces and organizational 
outcomes. Katowice: University of Economics in Katowice. 
33. Zwick, T. (2015). Training older employees: what is effective? International Journal of 
Manpower, 36(2), 136–150. 
 
 
 
 
 
 
 
 
 
 
 
 170 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
*** 
Tinkara Žabar is a junior researcher and a doctoral student in the Management program at the Faculty of 
Management, University of Primorska. She completed her undergraduate studies in pedagogy and obtained a 
master's degree in sociology, specializing in human resources, knowledge, and organizational management. Before 
joining the Faculty of Management, she gained practical experience in human resource management by working in 
HR departments of various organizations as a student, deepening her understanding of HR processes. 
She actively participates in the development of Erasmus+ Sport programs, through which she has attended 
numerous international conferences. Additionally, she has enhanced her professional competencies through 
further education, including the HR Starter program and the Leadership Academy. Her research focuses on the 
impact of artificial intelligence on employees, analyzing changes in the work environment as well as the challenges 
and opportunities brought by digital transformation. 
 
Tinkara Žabar je mlada raziskovalka in doktorska študentka na programu Management na Fakulteti za management 
Univerze na Primorskem. Dodiplomski študij je zaključila na področju pedagogike, magistrirala pa iz sociologije s 
specializacijo iz upravljanja človeških virov, znanja in organizacij. Pred zaposlitvijo na Fakulteti za management je 
kot študentka pridobivala praktične izkušnje na področju upravljanja človeških virov z delom v kadrovskih službah 
različnih organizacij, kjer je poglobila razumevanje kadrovskih procesov. 
Aktivno sodeluje pri razvoju programov Erasmus+ Sport, v okviru katerih se je udeležila številnih mednarodnih 
konferenc. Svoje strokovne kompetence je dodatno nadgrajevala z nadaljnjim izobraževanjem, med drugim v 
programih HR Starter in Akademija vodenja. Njeno raziskovalno delo je usmerjeno v proučevanje vpliva umetne 
inteligence na zaposlene, pri čemer analizira spremembe v delovnem okolju ter izzive in priložnosti, ki jih prinaša 
digitalna preobrazba. 
*** 
Aleksander Janeš is a Professor in the fields of Operational Management, Quality Management, and Business 
Excellence at the Faculty of Management, University of Primorska. As an experienced expert and researcher, he 
serves as the Program Director of the Master's in Management program, bringing over 29 years of professional 
experience. His research expertise and interests encompass various perspectives on project management systems 
and performance measurement, (green, blue, and sustainable) business models, business process management, 
as well as management tools in the fields of digitalization and process management, inclusive education and skills, 
and youth and media (https://orcid.org/0000-0001-5678-0737). 
 
Aleksander Janeš je profesor na področjih operativnega managementa, managementa kakovosti in poslovne 
odličnosti na Fakulteti za management Univerze na Primorskem. Kot izkušen strokovnjak in raziskovalec deluje tudi 
kot direktor magistrskega študijskega programa Management ter ima več kot 29 let delovnih izkušenj. Njegova 
raziskovalna strokovnost in interesna področja obsegajo različne vidike sistemov projektnega vodenja in merjenja 
uspešnosti, (zelene, modre in trajnostne) poslovne modele, management poslovnih procesov ter managerska 
orodja na področjih digitalizacije in procesnega managementa, vključujočega izobraževanja in razvoja kompetenc 
ter mladih in medijev. 
*** 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 171 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
Povzetek 
Izpopolnjevanje starejših zaposlenih v dobi umetne inteligence  
 
Raziskovalno vprašanje (RV): Kakšen je učinek novih tehnologij, s poudarkom na umetno inteligenco 
(UI), na potrebo po izpopolnjevanju starejših zaposlenih (50+ let). 
Namen: Namen raziskave je bil opraviti sistematični pregled literature dosedanjih raziskav s področja 
učinka UI na potrebe po izpopolnjevanju starejših zaposlenih. 
Metoda: Opravili smo sistematični pregled literature v šestih akademskih iskalnikih in sicer ProQuest, 
Emerald, Sage Journals, Springer, Research Gate ter v Google Učenjaku.  
Rezultati: Umetna inteligenca pomembno preoblikuje trg dela, saj zahteva nenehno prilagajanje 
novim spretnostim in znanju. Pomemben učinek ima UI na starejše zaposlene, ki so zaradi 
morebitnega pomanjkanja digitalnih veščin in občutljivosti na spremembe izpostavljeni večjim 
izzivom. V tem kontekstu sta izpopolnjevanje in dodatna izobrazba ključna mehanizma za 
zagotavljanje skladnosti spretnosti z zahtevami delovnega okolja in trga dela. Organizacije se morajo 
hitro prilagajati spreminjajočim se zahtevam z oblikovanjem kulture vseživljenjskega učenja, ki 
spodbuja starejše in ostale zaposlene k izpopolnjevanju. Ključno je, da izobraževalni programi 
temeljijo na specifičnih potrebah in izzivih, s katerimi se soočajo starejši zaposleni.   
Organizacija: Raziskava poudarja pomen izpopolnjevanja starejših zaposlenih v dobi umetne 
inteligence in organizacije spodbujanja k ustvarjanju kulture vseživljenjskega učenja, kot dela 
strateških usmeritev in ciljev organizacije. 
Družba: Pomen raziskave za družbo se odraža v vpogledu vključenosti vseh starostnih skupin v 
možnosti izpopolnjevanja znanja, veščin in odnosa do uporabe sodobnih tehnologij. Organizacije in 
družba sama je namreč nosilec socialne odgovornosti, da starejšim zaposlenim omogočijo uspešno 
vključevanje v delovno okolje v dobi UI.  
Originalnost: Raziskava naslavlja potrebo po izpopolnjevanju specifične starostne skupine v dobi UI, 
kjer sočasno osvetljuje pomen ustvarjanja kulture vseživljenjskega učenja v hitro se spreminjajočem 
svetu. 
Omejitve/nadaljnje raziskovanje: Pregled literature je bil omejen na šest javno dostopnih baz 
podatkov. V članku so bili starejši zaposleni obravnavani kot vse osebe v delovnem procesu starejše 
kot 50 let. Pri tem je potrebno izpostaviti, da se starejši zaposleni med seboj razlikujejo glede na 
izobrazbo, ekonomske, socialne in druge okoliščine. Primerno bi bilo, da bi učinek novih tehnologij 
raziskali tudi glede na omenjene okoliščine pri tej starostni skupini.  
 
Ključne besede: družba znanja, izpopolnjevanje, management izobraževanja, prekvalifikacija, 
starejši zaposleni, umetna inteligenca, vseživljenjsko učenje.  
 
 
  This work is licensed under a Creative Commons Attribution 4.0 International License.   
 This journal is published by Faculty of Organisation Studies in Novo mesto. 
 
 
 
 
 
 
 172 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
DOI: 10.37886/ip.2025.008 
 
 
Obstacles to the Implementation of Innovations 
in Start-Up Companies 
 
Nerman Ljevo1   , Sabina Šehić - Kršlak2  
1 Faculty for Management and Business Economy, Bosnia and Herzegovina, ljevo.nerman@gmail.com  
2 Center for Advanced Technologies, Bosnia and Herzegovina, ssabina83@yahoo.com 
 
 
Abstract 
Research Question (RQ): Start-up companies represent the support of the development of every 
economy. However, this is a more fragile part of the economy, because a large number of start-up 
companies fail in the first years of their development. The results of previous research indicate that 
there are several reasons for the failure of start-ups, and one of the key ones is the difficult application 
of innovations. In today's modern and turbulent times, in the embrace of globalization, innovations 
are an absolute must have. All companies that are not innovatively oriented are doomed to failure. 
Even large, multinational companies pay for the lack of innovation, so this is a signal to start-ups that 
investing in innovation is the most important thing in their budget. However, there are numerous 
challenges and obstacles in the application of innovations at start-ups. The main problem of the 
research is to detect what are the most important obstacles and challenges for the application of 
innovations in Bosnian start-ups, and by what methods it is possible to overcome them.  
Purpose: The main purpose of the research is to point out the obstacles and challenges faced by start-
up companies in Bosnia and Herzegovina and to give adequate recommendations for start-ups as well 
as all interest groups, how to eliminate them effectively, in order to make the economy and business 
operations of the company more propulsive and proactive. 
Method: In the theoretical explanation of the problem, sufficient information contained in previous 
research studies will be gathered, regarding the considered issue. Through empirical research, it will 
be determined what the most common types of obstacles are and challenges faced by start-ups in 
Bosnia and Herzegovina, what types of innovations are most often implemented in start-up 
companies in Bosnia and Herzegovina, and which methods and ways can be used to overcome the 
mentioned obstacles. The research will be done on a sample of 10 start-up companies in BiH. 
Results: The results of the research will show what are the obstacles and challenges in the 
implementation of innovations recorded in the sample of observed start-ups, what types of 
innovations are most often implemented in start-ups in Bosnia and Herzegovina, and how to most 
effectively overcome obstacles in the implementation of innovations in start-ups, with the aim of their 
better and more successful business.  
Organization: Through research of this type, managers of start-ups (or those who intend to become 
one) can more easily identify the basic obstacles in the implementation of innovations in their 
business, and use the proposed methods to overcome them, in order to make their business more 
stable and sustainable. 
Society: Increasing the stability of start-ups brings long-term benefits for the entire society. Socially 
responsible business of start-ups can contribute to a better distribution of resources in society, which 
is desirable for all participants in society, but also for the environment. 
Originality: Research of this type represents one of the rare research studies on start-ups in Bosnia 
and Herzegovina. It is pioneering research in the context of challenges and obstacles in the 
implementation of innovations at start-ups in Bosnia and Herzegovina. 
 
Received: 2025-05-22, revised: 2025-06-19, accepted: 2025-11-06  173 
as Original Research Paper 
 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
Limitations / further research: This study has several limitations, including a small sample size and 
difficulties in data collection. Future research is recommended to focus on the implementation of 
innovations across different industries in which start-up companies operate. 
 
Keywords: start-up, innovation, challenges, obstacles, implementation, recommendation, stability, 
overcome. 
 
 
 
1 Introduction 
Start-up companies play a crucial role in the development of any economy. They are important 
for job creation, increased competition, diversification, higher productivity, improved efficiency, 
and more. It can be said that start-ups are the backbone of any economy. Start-ups can also be 
described as drivers of innovation development. However, during their establishment, growth, 
and expansion, start-ups are often insufficiently recognized by the very structures that should 
provide them with better and more sustainable market conditions. As a result, start-up founders 
face numerous obstacles that limit their growth and development and complicate their survival 
in the market. These obstacles are particularly related to the innovation implementation process.  
First and foremost, it is necessary to identify the key barriers start-ups face in the context of 
innovation implementation and then work to minimize them. This paper presents research in the 
field of entrepreneurship and aims to highlight the most common barriers that start-ups 
encounter in the innovation implementation process—both generally and specifically in Bosnia 
and Herzegovina—and explore the methods and strategies that can be used to overcome these 
barriers. Additionally, the paper seeks to identify the key stakeholders who can contribute to 
minimizing the challenges faced by start-ups. 
The main problem of the research is the insufficient identification and understanding of the 
barriers that start-up companies face in the process of innovation implementation, particularly 
in the context of Bosnia and Herzegovina. Although start-ups are widely recognized as crucial 
drivers of innovation and economic development, they often encounter numerous challenges 
and barriers during their establishment, growth, and expansion. Therefore, this research in the 
field of entrepreneurship seeks to explore the nature and impact of these barriers, examine the 
strategies and methods used by start-ups to overcome them, and identify the key stakeholders 
who can contribute to improving the conditions under which start-ups operate. The ultimate goal 
is to generate findings that can help inform more supportive policies and practices aimed at 
fostering innovation and entrepreneurship in the region. 
 
 
 
 174 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
2 Theoretical framework 
Entrepreneurship is crucial for the growth of any economy. It creates job opportunities, supports 
rural development, drives technological progress, increases national income, promotes 
industrialization, and boosts exports (Sinha, 2023). 
The survival of startups is closely tied to innovation. Over time, technological innovation has 
attracted growing attention from academics, businesses, industries, entrepreneurs, and 
policymakers, all of whom have increasingly focused on building an entrepreneurial ecosystem 
that can support and integrate innovation (Nicolau & Bărbulescu, 2024). According to Fakhri and 
Bahoussa (2014), since the time of Schumpeter, innovation has been closely associated with 
startups and involves a range of factors such as technology, nature, society, government, and the 
broader community. All these factors can—and do—have an impact on the implementation of 
innovations by startups.  
Sopjani (2019) argues that innovations implemented by startups contribute to positive social 
transformation. Therefore, all stakeholders involved in the innovation implementation process 
should work to ensure that it is carried out as efficiently as possible. Research conducted by 
Wetzel and Eiche (2024) on a sample of 152 startups shows that, at present, the most important 
factors for startups are developing an effective sales and customer service system, 
implementing financial planning, and adopting a sustainable pricing and costing model. 
Alawamleh et al. (2023), based on a sample of startups in Jordan, found that women encounter 
more innovation-related challenges than men. These challenges often relate to limited access to 
funding and a general lack of investment in startups. 
According to Banka et al. (2024), innovation challenges in startups often stem from a lack of 
understanding by investors and the high risk of business failure associated with implementing 
innovations. 
Wedajo and Kakuze (2020) point out that fear of digitalization or a lack of knowledge about digital 
tools can also pose barriers to innovation in startups. Supporting this view, Saksonova and 
Kuzmina-Merlino (2017) argue that startup founders must be ready to manage all business 
obligations digitally, including financial operations. Silva Nunes et al. (2021) emphasize the need 
to reform laws and public institution practices so they are fully aligned with the needs of 
innovation implementation in startups. This, they argue, is the only way to foster a strong 
entrepreneurial culture and spirit. Rajiani et al. (2023) go a step further, suggesting that 
governments should assist startups in implementing new technologies, thereby facilitating 
innovation development. 
Carle (2024) finds that smaller startups tend to offer better innovations than larger ones due to 
their agility and responsiveness to market changes. Based on research conducted on tech park 
  
 175 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
startups in France, she concludes that a significant threat to startups arises when they either fail 
to innovate or are unable to effectively implement or commercialize their innovations. Venczel et 
al. (2024) argue that startups must have a clear and precise methodology—essentially, a guiding 
algorithm—for implementing each new innovation. 
Kask and Linton (2025) highlight that knowledge is a critical factor in innovation implementation. 
Education and expertise can make the difference between the success and failure of innovative 
startup ventures. On the other hand, Kuckertz et al. (2020) point out the importance of 
motivation: individuals involved in innovation must be properly motivated, as motivation greatly 
influences whether the innovation process aligns with organizational goals. Another critical 
element is talent. According to Brunetti et al. (2020), talent is necessary for successful innovation 
implementation. A lack of talent can pose a serious challenge for startups. 
Hvolkova et al. (2019) suggest models for overcoming barriers to innovation implementation. 
One proposed model involves categorizing barriers based on their significance and focusing on 
eliminating the most critical ones to accelerate and intensify innovation efforts. This approach 
requires the involvement of all stakeholder groups connected to startups. Finally, 
Kusumaningtyas et al. (2023) propose a five-step model to facilitate innovation implementation 
in startups. The first step of this model emphasizes identifying profit opportunities from 
innovation—recognizing from the outset the potential to recover the time, effort, and resources 
invested. This return should come directly from the market. 
In the next Table (Table 1), it can be seen the main barrieres to innovation implementation. 
  
  
 176 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
Table 1  
Mainbarrieres to innovation implementation 
Category of 
Number Description of Barrier Sources 
Barrier 
1 Financial Barriers Limited access to funding and lack of Alawamleh et al. (2023), Banka 
investment, especially for women in start- et al. (2024) 
ups. 
2 Knowledge and Lack of knowledge about digital tools and Wedajo & Kakuze (2020), Kask & 
Education insufficient education on innovation Linton (2025) 
management. 
3 Motivation and Lack of motivation and talent necessary for Kuckertz et al. (2020), Brunetti 
Talent successful innovation implementation. et al. (2020) 
4 Institutional Absence of adequate laws and public Silva Nunes et al. (2021) 
Barriers institution practices aligned with innovation 
needs. 
5 Fear of Resistance or fear of digitalization that Wedajo & Kakuze (2020), 
Digitalization hinders the use of digital technologies in Saksonova & Kuzmina-Merlino 
business. (2017) 
6 Inadequate Lack of clear and precise methods or Venczel et al. (2024) 
Methodology algorithms for innovation implementation. 
7 Risk of Business High risk of failure associated with Banka et al. (2024) 
Failure innovation implementation discourages 
investors. 
8 Lack of Market Insufficient ecosystem support, including Fakhri & Bahoussa (2014), 
Support lack of relevant stakeholders to support Rajiani et al. (2023), Hvolkova et 
innovations. al. (2019) 
 
The basic assumption of the research is that financial barriers are the most common form of 
obstacle for startups in Bosnia and Herzegovina. This will be elaborated on further in the paper. 
3 Method 
This study was conducted as an exploratory cross-sectional study with a mixed quantitative-
qualitative design. The primary aim of the research was to identify the obstacles and challenges 
faced by start-up companies in Bosnia and Herzegovina during innovation implementation and 
to provide recommendations for effectively overcoming these barriers, thus making the business 
operations and economy of these companies more dynamic and proactive. Data were collected 
from a sample of 10 start-up companies in Bosnia and Herzegovina. The inclusion criteria were 
that the company must be a start-up (i.e., no older than one year), located in Bosnia and 
Herzegovina, and actively engaged in some form of innovation. The sample was selected using 
purposive sampling due to the specific nature of the population. The data collection instruments 
included an online questionnaire and semi-structured interviews with start-up founders. The 
questionnaire contained questions about the industry sector, number of employees, year of 
establishment, type of innovation implemented, perceived importance of innovation, and 
  
 177 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
obstacles faced during the innovation implementation process. The interviews provided 
additional qualitative insights into participants’ experiences and challenges. The sampling 
process was defined by clear inclusion and exclusion criteria, excluding companies that did not 
meet the age requirement or were not actively innovating. Given the sample size and approach, 
the study’s limitations include potential limited generalizability of the results. For data analysis, 
statistical tools MS Excel and SPSS were used. Quantitative data were processed using 
descriptive statistics, while qualitative interview data were analyzed using content coding 
techniques to identify key themes and patterns. Ethical considerations included obtaining 
informed consent from all participants, ensuring confidentiality and anonymity of respondents. 
The study was conducted following ethical guidelines for research involving human subjects, and 
institutional approval was obtained where applicable. 
4 Results 
At the very beginning, the data related to the sample are presented in Table 2. 
Table 2  
Information about sample 
Number Industry Number of employers Year of establishment 
1 Textile 2 2024 
2 Law 1 2024 
3 It 2 2024 
4 It 3 2024 
5 It 6 2024 
6 It 5 2024 
7 It 2 2024 
8 It 2 2024 
9 Food Industry 2 2024 
10 Cosmetics 1 2024 
 
As can be seen, the sample included 10 start-ups, all founded in 2024 (i.e., none are older than 
one year). The majority of the companies operate in the field of information technology (60%), 
while the remaining companies belong to the textile sector, law, food industry, and cosmetics. 
The highest number of employees was recorded in one IT company (6 employees), while the 
lowest was observed in two companies, each employing only one person. 
The types of innovations being implemented are presented in Figure 1. 
 
 
 
  
 178 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
Figure 1  
Types of innovations 
 
 
 
The research results indicate that the majority of startups in the sample are implementing 
technological innovations, 30 % are engaged in process innovations, while only 10 % are 
introducing product innovations. It is also important to note that the respondents were asked 
whether they consider innovation important for their business, and whether they find it profitable 
and effective for use in the market. Of all respondents, 100 % stated that they believe their 
innovation is very important for their business, that it is very profitable, and that it is effective for 
broader market use. However, when it comes to the need for further development of their 
innovations, 60 % of respondents believe that additional work is required. 
Figure 2 presents whether the startups in the sample have previously faced any issues in the 
process of implementing innovations. 
  
  
 179 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
Figure 2  
Obstacles in implementing innovations 
 
 
Based on the previous graph, it can be concluded that all 10 startups from Bosnia and 
Herzegovina included in the sample have encountered some form of obstacle during the 
implementation of innovations. 
The types of obstacles are shown in Figure 3. 
Figure 3  
Obstacles to the implementation of innovations 
 
 
Based on the previous graph, it can be concluded that the majority of startups face a lack of 
financial resources when implementing their innovations. It is evident that every innovation 
requires substantial investment, making financial issues the most common and significant 
challenge among the companies in the sample. Five respondents reported that the biggest 
problem in implementing innovations was a lack of financial support. This was followed by a lack 
  
 180 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
of knowledge and lack of vision, each mentioned by two respondents, while one respondent 
indicated that the greatest problem was insufficient planning. 
Through the analysis of the proposed statements, the aim was to identify potential solutions for 
overcoming the obstacles to innovation implementation. Respondents rated the statements on 
a scale of 1 to 5. The analysis is presented in Table 3. 
Table 3  
Statements 
 
No Statements Respondents' ratings Mean 
1 Encouraging creativity (hackathons, brainstorming 4 4 4 4 4 4 4 4 4 3 3,9 
sessions) 
2 Insisting on building a prototype  4 4 4 4 4 4 4 4 3 1 3,6 
3 Introduction of agile methods (Scrum, Kanban)  4 4 4 4 4 4 4 4 4 4 4,0 
4 Crowdsourcing and crowdfunding  4 4 4 4 4 4 4 5 5 5 4,3 
5 Investing in technology, AI, digitalization  4 4 4 4 4 4 4 5 5 5 4,3 
6 Mentoring and investment support 4 4 4 4 4 4 5 5 5 5 4,4 
 
Based on the previous table, it is evident that startups in Bosnia and Herzegovina most need 
knowledge and investment, that is, mentoring support and support in the context of finding 
certain sources of financing. On the other hand, they least need anyone to motivate them to 
implement innovations and work on them, they are already motivated enough to offer their 
innovations to the market.  
5 Discussion 
The study included a sample of ten start-ups from Bosnia and Herzegovina, all established in 
2024, with the majority operating in the field of information technology. The analysis reveals that 
most of these start-ups are implementing technological innovations, while process and product 
innovations are less represented. Notably, all respondents emphasized that innovation is of vital 
importance for their business, describing it as both profitable and effective for broader market 
application. Nevertheless, 60% of the respondents expressed the need for further development 
of their innovations. All start-ups in the sample reported encountering certain obstacles during 
the innovation implementation process. The most frequently cited barrier was the lack of 
financial resources, reported by half of the respondents. This was followed by a lack of knowledge 
and a lack of vision (each mentioned by two respondents), and insufficient planning (mentioned 
by one respondent). These findings suggest that financial constraints represent a predominant 
challenge for young companies seeking to innovate. Furthermore, respondents identified the 
greatest need for support in the form of mentoring and access to investment or financing 
opportunities. In contrast, motivational support was deemed least necessary, as participants 
reported being highly self-motivated to develop and introduce their innovations to the market. 
  
 181 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
The presented results indicate that startups are not deprived or protected from obstacles when 
implementing innovations. Each of the startups in the sample had some form of obstacles when 
implementing their innovations. It is important to note that in most cases (as much as 50%) the 
lack of finance marked the main obstacle in the implementation of innovations. Similar results 
were recorded by Mohnen et al. (2008) and García-Quevedo et al. (2018). It is important to point 
out that financial problems are not only characteristic of startups from Bosnia and Herzegovina, 
but also in other, much more developed countries. Based on these results, it can be concluded 
that the basic assumption of the research is confirmed. On the other hand, startups need 
professional and mentoring help, i.e. additional knowledge in the field of management, in order 
to successfully run their organizations. Through the conversation with startups, certain ideas and 
directions were reached on how to improve the entrepreneurial climate in Bosnia and 
Herzegovina. Some of the proposals were: Grants and subsidies (on the basis of which a certain 
amount of financial resources would be accumulated that would be used to invest in 
innovations), Technological parks and incubators (there is a lack of space in Bosnia and 
Herzegovina that would support startups - in terms of space but also in financial and mentoring 
terms, and the construction of technological parks and incubators is what should be done by the 
government in the coming period); Creating a favorable regulatory environment (it is necessary 
to enable easier registration for startups, in order to support the creation of an entrepreneurial 
climate); Support in education and training. Creating a community and network (there is a need 
to network startups, which can be achieved precisely through the previously mentioned 
technology parks and incubators); Support in entering the foreign market (assistance in the 
internationalization of business for startups must be the vision of the governing structure), Legal 
aid and Business angels (legal aid is necessary for all startups, as well as connections with 
business angels, which can be achieved through the organization of certain competitions, 
presentations, etc.). 
It should be noted that the current institutional support for startups in Bosnia and Herzegovina is 
insufficient, and that it is necessary to work on this area. The difficulties that startups encounter, 
which are recorded in the work, can be overcome institutionally and facilitate their operations. 
The results and recommendations previously given can be a significant input to government 
institutions on how to act to increase the number of startups and make the business sector of 
Bosnia and Herzegovina more developed. 
6 Conclusion 
In this part, the basic conclusions will be presented. First of all, it is important to emphasize that 
the process of analyzing secondary and subsequently primary data aimed to demonstrate the 
importance of start-ups for any economy, particularly highlighting the significance of the 
innovations they generate. Accordingly, the analysis sought to identify the key obstacles to 
innovation and underscore the importance of minimizing these barriers in order to enable start-
  
 182 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
ups to operate at their full creative potential. The research found that the lack of financial 
resources is one of the key barriers to innovation, and it is important that the government and all 
interest groups provide access to financial resources to startups, whose growth is needed by the 
economy. Something like this is possible through grants, subsidies, but also through cooperation 
with business angels, enabling space through technology parks and incubators. In addition to the 
above, the startups mentioned the need to work with mentors, as well as the need for knowledge 
and education, because this factor can be crucial when implementing innovations. After the 
basic categories of obstacles have been defined, it is important to eliminate each of the existing 
ones, in order to make startups work better and more efficiently. The results of this research can 
be used by the government and its institutions in Bosnia and Herzegovina, but also by the wider 
community, which can make a step forward in creating an entrepreneurial environment, and 
improve the status of startups in general. It is a process that benefits everyone. 
 
References 
1. Alawamleh, M., Francis, Y. H., & Alawamleh, K. J. (2023). Entrepreneurship challenges: The 
case of Jordanian start-ups. Journal of Innovation and Entrepreneurship, 12(21). 
https://doi.org/10.1186/s13731-023-00286-z 
2. Banka, M., Chmiel, N., Kostrzewski, M., Marczewska, M., Kowalski, A. M., Sedkiewicz, K., & 
Salwin, M. (2024). Understanding corporate concerns: Barriers and challenges in 
corporate–start-up collaboration. Journal of Open Innovation: Technology, Market, and 
Complexity, 10(4), 100388. https://doi.org/10.1016/j.joitmc.2024.100388 
3. Brunetti, F., Matt, D. T., Pedrini, G., & Orzes, G. (2020). Digital transformation challenges: 
Strategies emerging from a multi-stakeholder approach. The TQM Journal, 32(4), 697–724. 
https://doi.org/10.1108/TQM-12-2019-0300 
4. Carle, A. (2024). Implementation challenges of innovation policies fostering sustainability: 
Evidence from a French public grant for technological startups. Journal of Innovation 
Economics & Management, (43)(1). 
5. Fakhri, S., & Bahoussa, A. (2014). Obstacles of innovation among the entrepreneur: An 
empirical study. International Journal of Innovation and Applied Studies, 9(1), 393–400. 
http://www.ijias.issr-journals.org/ 
6. García-Quevedo, J., Segarra-Blasco, A., & Teruel, M. (2018). Financial constraints and the 
failure of innovation projects. Technological Forecasting and Social Change, 127, 127–140. 
https://doi.org/10.1016/j.techfore.2017.09.027 
7. Hvolkova, L., Klement, L., Klementova, V., & Kovalova, M. (2019). Barriers hindering 
innovations in small and medium-sized enterprises. Journal of Competitiveness, 11(2), 51–
67. https://doi.org/10.7441/joc.2019.02.04 
  
 183 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
8. Kask, J., & Linton, G. (2025). Navigating the innovation process: Challenges faced by deep-
tech startups. In Contemporary Issues in Industry 5.0. 
9. Kuckertz, A., Brändle, L., Gaudig, A., Hinderer, S., Morales Reyes, C. A., Prochotta, A., 
Steinbrink, K. M., & Berger, E. S. C. (2020). Startups in times of crisis – A rapid response to 
the COVID-19 pandemic. Journal of Business Venturing Insights, 13, e00169. 
https://doi.org/10.1016/j.jbvi.2020.e00169 
10. Kusumaningtyas, A., Bolo, E., Istianah, I., Chua, S., Wiratama, M., & Tirdasari, N. L. (2021). 
Why start-ups fail: Cases, challenges, and solutions. In Proceedings of the Conference 
Towards ASEAN Chairmanship 2023 (T-A-C 23 2021). Advances in Economics, Business and 
Management Research, 198. 
11. Mohnen, P., Palm, F. C., Schim van der Loeff, S., & Tiwari, A. (2008). Financial constraints 
and other obstacles: Are they a threat to innovation activity? De Economist, 156(2), 201–
214. https://doi.org/10.1007/s10645-008-9089-y 
12. Nicolau, C., & Bărbulescu, O. (2024). Innovation challenges and benefits in high-tech start-
ups: A quantitative analysis of university student entrepreneurs in Romania. Amazonia 
Investiga, 13(82), 222–235. https://doi.org/10.34069/AI/2024.82.10.18 
13. Nunes, A. K. da S., Morioka, S. N., & Bolis, I. (2022). Challenges of business models for 
sustainability in startups. RAUSP Management Journal, 57(4), 382–400. 
https://doi.org/10.1108/RAUSP-10-2021-0216 
14. Rajiani, I., Kot, S., Michałek, J., & Riana, I. G. (2023). Barriers to technology innovation 
among nascent entrepreneurs in deprived areas. Problems and Perspectives in 
Management, 21(3). 
15. Saksonova, S., & Kuzmina-Merlino, I. (2017). Fintech as financial innovation – The 
possibilities and problems of implementation. European Research Studies Journal, 20(3A), 
961–973. 
16. Sinha, N. (2023). Challenges and opportunities faced by innovative entrepreneurs in India. 
International Journal of Innovative Research in Management and Political Science (IJIRMPS), 
11(2). 
17. Sopjani, X. (2019). Challenges and opportunities for startup innovation and 
entrepreneurship as tools towards a knowledge-based economy: The case of Kosovo 
(Master's thesis). Rochester Institute of Technology. RIT Digital Institutional Repository. 
18. Venczel, T. B., Berényi, L., & Hriczó, K. (2024). The project and risk management challenges 
of start-ups. Acta Polytechnica Hungarica, 21(2). 
19. Wedajo, B. T., & Kakuze, H. (2020). Barriers in digital startup scaling: A case study of 
Northern Ethiopia (Master’s thesis). Umeå University. 
20. Wetzel, T., & Eiche, J. (2024). Challenges of start-ups—An analysis of individually tailored 
recommendations based on the development phases, branches, business models and 
founding teams. Open Journal of Business and Management, 12, 1556–1585. 
https://doi.org/10.4236/ojbm.2024.123084 
  
 184 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
Povzetek 
Izzivi pri uvajanju inovacij v zagonskih podjetjih 
 
Raziskovalno vprašanje (RV): Zagonska podjetja predstavljajo pomembno podporo razvoju vsakega 
gospodarstva, hkrati pa so tudi njegov bolj ranljiv del, saj veliko zagonskih podjetij propade že v prvih 
letih delovanja. Rezultati dosedanjih raziskav kažejo, da obstaja več razlogov za neuspeh zagonskih 
podjetij, med ključnimi pa je oteženo uvajanje inovacij. V sodobnem, turbulentnem času in v razmerah 
globalizacije so inovacije nujen pogoj za obstoj in razvoj podjetij. Podjetja, ki niso inovativno 
usmerjena, so dolgoročno obsojena na propad. Tudi velika multinacionalna podjetja občutijo 
posledice pomanjkanja inovativnosti, kar je jasen pokazatelj, da so vlaganja v inovacije za zagonska 
podjetja izjemnega pomena. Kljub temu pa se pri uvajanju inovacij pojavljajo številni izzivi in ovire. 
Glavni problem raziskave je ugotoviti, katere so najpomembnejše ovire in izzivi pri uvajanju inovacij v 
zagonskih podjetjih v Bosni in Hercegovini ter s katerimi metodami jih je mogoče premagati. 
Namen: Glavni namen raziskave je opozoriti na ovire in izzive, s katerimi se srečujejo zagonska 
podjetja v Bosni in Hercegovini, ter podati ustrezna priporočila tako zagonskim podjetjem kot tudi 
vsem ključnim deležnikom, kako te ovire čim bolj učinkovito odpraviti, da bi bilo gospodarstvo bolj 
propulzivno in proaktivno. 
Metoda: Pri teoretični obravnavi problematike bodo uporabljeni podatki iz dosedanjih raziskav s tega 
področja. Z empirično raziskavo bo ugotovljeno, katere so najpogostejše vrste ovir in izzivov, s katerimi 
se soočajo zagonska podjetja v Bosni in Hercegovini, katere vrste inovacij se v teh podjetjih 
najpogosteje uvajajo ter katere metode in načine je mogoče uporabiti za premagovanje teh ovir. 
Raziskava bo izvedena na vzorcu desetih zagonskih podjetij v Bosni in Hercegovini. 
Rezultati: Rezultati raziskave bodo pokazali, katere ovire in izzivi se pojavljajo pri uvajanju inovacij v 
opazovanih zagonskih podjetjih, katere vrste inovacij se v zagonskih podjetjih v Bosni in Hercegovini 
najpogosteje izvajajo ter kako najučinkoviteje premagovati ovire pri uvajanju inovacij z namenom 
izboljšanja njihovega poslovanja. 
Organizacija: S tovrstnimi raziskavami lahko vodje zagonskih podjetij oziroma tisti, ki to nameravajo 
postati, lažje prepoznajo temeljne ovire pri uvajanju inovacij v svoje poslovanje ter uporabijo 
predlagane metode za njihovo premagovanje, kar prispeva k stabilnejšemu in trajnostnemu 
poslovanju. 
Družba: Povečanje stabilnosti zagonskih podjetij prinaša dolgoročne koristi za celotno družbo. 
Družbeno odgovorno poslovanje zagonskih podjetij lahko prispeva k boljši porazdelitvi virov v družbi, 
kar je koristno za vse družbene deležnike ter tudi za okolje. 
Originalnost: Tovrstna raziskava predstavlja eno redkih raziskav o zagonskih podjetjih v Bosni in 
Hercegovini. Gre za pionirsko raziskavo na področju izzivov in ovir pri uvajanju inovacij v zagonskih 
podjetjih v tej državi. 
Omejitve/nadaljnje raziskovanje: Ta raziskava ima več omejitev, med katerimi sta majhna velikost 
vzorca in težave pri zbiranju podatkov. Nadaljnje raziskave naj se osredotočijo na uvajanje inovacij v 
različnih panogah, v katerih delujejo zagonska podjetja. 
 
Ključne besede: zagon, inovativnost, izzivi, ovire, implementacija, priporočila, stabilnost, 
premagovanje. 
 
 
 
 
 
 
  
 185 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
*** 
Nerman Ljevo je redni profesor (docent) na Fakulteti za management in poslovno ekonomijo Univerze v Travniku. 
Doktorski študij je zaključil na Ekonomski fakulteti Univerze v Tuzli, smer Management in organizacija. Objavil je več 
deset znanstvenih člankov in dve knjigi. 
 
Nerman Ljevo is an Associate Professor at the Faculty of Management and Business Economics of the University of 
Travnik. He completed his doctoral studies at the Faculty of Economics, University of Tuzla, specializing in 
Management and Organization. He has published several dozen scientific articles and two books. 
 
*** 
Sabina Šehić - Kršlak je profesorica na Fakulteti za management in poslovno ekonomijo Univerze v Travniku, 
Fakulteti za javno upravo Univerze v Sarajevu in uslužbenka Centra za napredne tehnologije Kantona Sarajevo. 
Doktorski študij je zaključila na Ekonomski fakulteti Univerze v Banja Luki, smer Management in organizacija. 
Objavila je več deset znanstvenih člankov in tri knjige. 
 
Sabina Šehić - Kršlak is a Professor at the Faculty of Management and Business Economics of the University of 
Travnik, the Faculty of Public Administration of the University of Sarajevo, and an employee of the Canton Sarajevo 
Center for Advanced Technologies. She completed her doctoral studies at the Faculty of Economics of the University 
of Banja Luka, specializing in Management and Organization. She has published several dozen scientific articles and 
three books. 
*** 
 
 
 
  This work is licensed under a Creative Commons Attribution 4.0 International License.   
 This journal is published by Faculty of Organisation Studies in Novo mesto. 
 
 
 
 
 
 
 
 
 
 
 
  
 186 
ISSN 2463-9281
Izid publikacije je finančno podprla ARIS iz naslova razpisa za
sofinanciranje domačih znanstvenih periodičnih publikacij.
The journal is subsidised by the Slovenian Research and Innovation Agency.
GLAVNA IN ODGOVORNA UREDNICA / 
EDITOR IN CHIEF A N N M A R I E  G O R E N C  Z O R A N
SOUREDNICA / 
ASSOCIATE EDITOR N A D I A  M O L E K
UREDNIŠKI ODBOR /
EDITORIAL BOARD
Boris Bukovec, Faculty of Organisation Studies in Novo mesto, Slovenia
Alois Paulin, Technical University Vienna, Austria
Juraj Marušiak, Slovak Academy of Science, Slovakia
Mario Ianniello, Udine University, Italy
Anisoara Popa, Danubius University, Romania
Raluca Viman-Miller, University of North Georgia, Georgia, USA
Anna Kołomycew, Rzeszów University, Poland
Jurgita Mikolaityte, Siauliai University, Lithuania
Patricia Kaplanova, Faculty of Organisation Studies in Novo mesto, Slovenia
Laura Davidel, Univeristy of Lorraine, France
Ana Železnik, Ljubljana University, Slovenia
Marko Vulić, Information Technology School - ITS ComTrade, Serbia
Vita Jukneviciene, Siauliai University, Lithuania
Mitja Durnik, Ljubljana University, Slovenia
Anca-Olga Andronic, Spiru Haret University, Romunija
Razvan-Lucian Andronic, Spiru Haret University, Romunija
Tine Bertoncel, Faculty of Organisation Studies in Novo mesto, Slovenia
Nadia Molek, Faculty of Organisation Studies in Novo mesto, Slovenia
Maja Meško, University of Maribor, Faculty of Organizational Sciences, Slovenia
Agnieszka Wedel-Domaradzka, Faculty of Law nad Economics, Kazimierz Wielki University,
Bydgoszcz, Poland
Armand Faganel, University of Primorska, Faculty of Management, Slovenia
Elżbieta Roszko-Wójtowicz, University of Lodz, Poland
Gorazd Justinek, New University, Faculty of Government and European Studies
Damyana Bakardzhieva, Anwar Gargash Diplomatic Academy, United Arab Emirates (UAE)
Bashar H. Malkawi, University of Arizona James E. Rogers College of Law, USA
Kodzo Alabo, University of Ghana, Salt University College, Ghana
Robert Mudida, Strathmore University Business School, Nairobi, Kenia
Juan Carlos Radovich, Universidad de Buenos Aires, Argentina
Anurag Hazarika, Tezpur Central University, Assam, India & 21st century Open University, USA 
Naslov uredništva / Editorial address: 
Fakulteta za organizacijske študije v Novem mestu
Ulica talcev 3
8000 Novo mesto, Slovenija
 
Licenca CC BY-SA 4.0 omogoča prosto uporabo,  pr i lagajanje in distr ibuci jo dela,  tudi  v
komercialne namene,  pod pogojem, da se pr izna avtorstvo in ohrani  enaka l icenca pri
del jenih predelavah.  To zagotavl ja odprtost in šir jenje vsebine pod enakimi pogoj i .
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
 
DOI: 10.37886/ip.2025.009 
 
 
AI is not a Tool 
The Impact of Growing AI Agency on the Future of Work 
 
Alexander van Biezen1    
1 Arcadia University, Belgium, alexander.vanbiezen@arcadiascholen.be  
 
 
Abstract 
Research Question (RQ): What are the underlying philosophical assumptions shaping current 
perceptions of artificial intelligence (AI) as a mere tool, and how do these assumptions influence our 
understanding of AI’s growing agency and its potential impact on the future of work? 
Purpose: The paper aims to critically examine the widespread assumption that AI systems remain 
passive instruments entirely under human control. It explores how emerging forms of AI agency—
understood as autonomous or semi-autonomous decision-making capacities—challenge this notion 
and what implications this shift entails for human labour, ethics, and social stability. 
Methods: The study adopts a philosophical and conceptual methodology grounded in the 
philosophy of mind and the philosophy of science. It draws on classical thought experiments 
(Searle’s Chinese Room, Jackson’s Mary, Penrose’s arguments on non-algorithmic consciousness) 
and integrates recent interdisciplinary debates on AI agency, autonomy, and consciousness. The 
analysis is based on a critical literature review combining philosophical, technological, and socio-
political sources. 
Results: Findings indicate that the assumption of AI as a “dumb tool” no longer holds. Evidence of 
growing AI autonomy demonstrates that decision-making processes once reserved for humans are 
increasingly being delegated to machines. This outsourcing of human agency risks creating social 
and ethical blind spots, potentially leading to unequal labour transformations and governance 
challenges. However, a managed transition toward human–AI cooperation could foster innovation 
and inclusion if grounded in ethical oversight and policy regulation. 
Organization: For organizations, the study highlights the need to anticipate shifts in work structures 
and decision-making processes caused by AI systems with growing agency. It encourages managers 
and policymakers to design governance frameworks that maintain human oversight while enabling 
responsible collaboration with AI. 
Society: At the societal level, the research underlines the urgency of open policy debates and ethical 
reflection on AI regulation. Addressing the implications of AI autonomy is essential to preserve 
human agency, democratic accountability, and social justice in the digital era. 
Originality: The article contributes to bridging philosophical inquiry and socio-technical analysis by 
reframing AI not merely as a technological tool but as an emerging actor in human decision-making 
systems. It advances the concept of “AI agency” as a key lens for understanding the transformation 
of work. 
Limitations / Further research: The study is conceptual and does not include empirical data. Future 
research should investigate how organizations and workers experience AI agency in practice, 
possibly through ethnographic or organizational case studies, and explore policy instruments 
capable of mitigating risks related to automation and technocratic governance. 
 
Keywords: artificial intelligence, AI agency, AI consciousness, workforce, future, philosophy of science, 
philosophy of mind. 
 
 
Received: 2025-06-24, revised: 2025-07-02, accepted: 2025-11-05  188 
as Original Research Paper 
 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
1 Introduction 
Sometimes reality catches up with us faster than expected. The past few years, each time when 
I mentioned the possibility of AI systems developing ‘agency’ to my students, I was met with 
blank stares or frowning. Today, ‘AI agent’ has become a new buzzword overnight.  At the 
beginning of March 2025, a report from The Information announced that OpenAI, one of the 
leading AI research organizations, may be planning to charge up to $20,000 per month for 
specialized AI ‘agents’ (Palazzo & Weinberg, 2025).  The prices vary from $2,000 a month for an 
AI agent at the level of a ‘high-income knowledge worker’ (Wiggers, 2025), a software developer 
agent will cost about $10,000 a month, and at the top we find a ‘Phd-level research’ (Edwards, 
2025) agent which will cost you no less than $20,000 a month. 
The etymological origin of 'agency' stems from the Latin verb agere which means “to do”, “to 
act”. The word 'agent' stems from the present participle agens, agentis, “one who acts” or “one 
who does an act” (i.e. an agent). The topic of agency has a much longer history in the field of 
ethics, predating the advent of AI. Giving a precise definition of what characterizes 'agency' is 
quite a challenging task for philosophy, as it is related to intentional action and goal-directed 
behavior and, eventually, to the intricate philosophical question of what it means to be a 
person.  
According to a recent study from PULSE (Program on Understanding Law, Science, and 
Evidence) from UCLA School of Law, there is still a significant lack of agreement on the 
definition of agency and, as a result, there is still no consensus whether or not it is even 
possible to consider AI systems as agents (Newman et al., 2025). “(…) Depending on the 
perspective and definition (…) the agency of AI could be controversial, unimaginable, or an 
unquestionable truth (...)” (Newman et al., 2025). Especially the question whether or not an AI 
system can be considered as a goal-oriented entity remains controversial (Newman et al., 
2025).  
For our discussion at hand, we take the option of slipping through the horns of this dilemma. As 
a working definition of 'agency' with regard to AI in this article, we simply mean that some AI 
systems are getting to a whole new level of decision-making capabilities and autonomous 
action. Whether or not these current AI systems can really be said to have their own goals and 
are truly goal-oriented is not the key question in this respect. What matters here for our 
discussion at hand is that these AI systems are acquiring a level of autonomous decision-
making which makes it possible and tempting for us, humans, to transfer a growing part of 
human decision-making to these systems. 
Be that as it may, one thing’s for sure: developments in AI are making giant leaps at an ever-
increasing pace. In September 2024, the well-known historian Yuval Noah Harari published 
Nexus on information networks from early history to AI today. In Nexus, Harari warns us: “AI 
 
 
 189 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
isn’t a tool – it’s an agent” (Harari, 2024, p. XXII), meaning that AI is capable of processing 
information all by itself, and thereby has the capacity to replace humans in the making of 
decisions (Harari, 2024, p. XXII). Harari was heavily criticized for this warning, being put away by 
some as a doomsday prophet. For instance, Don Lim, a seasoned IT specialist and formerly 
Chief AI Developer at Visdex, was very quick to reproach Harari in his article Why Yuval Noah 
Harari’s AI Doomsday Prophecies Are Misleading (Lim, 2024) that he is being both alarmist and 
misinformed: 
“He also argues “AI is not a tool, but an agent.” The current AI systems are far 
from autonomous entities; they are tools created, monitored, and controlled 
by humans. The idea that AI will become a self-governing force beyond 
human control is closer to science fiction than reality.” (Lim, 2024)  
Some reproached him that his book Nexus is “based on shallow scholarship” (Ferguson, 2025), 
or even bluntly suggested that the real lesson we can learn from Harari is that “there’s an 
incredible amount of money to be made with doomsday predictions” (Foreman, 2024). 
But even Geoffrey Hinton, the so-called godfather of AI, who shared the 2024 Nobel Prize in 
Physics with John Hopfield for their foundational discoveries and inventions that enable 
machine learning with artificial networks, warns us that “AI systems may be more intelligent 
than we know and there’s a chance the machines could take over” and “we’re moving into a 
period when for the first time ever we may have things more intelligent than us” (Pelley, 2024). 
He even left Google a year before, in May 2023, precisely because of his concerns about the 
many risks of AI (Douglas Heaven, 2023).   
If even the very brightest and the most well-versed among us in the field of AI yield warnings 
about the rapid developments in AI and its possible unforeseen devastating consequences, 
maybe it is time to listen to what they have to say and, at least, to postpone our judgement, even 
if only for a moment. 
Vincent Ginnis, professor of mathematics, physics and artificial intelligence at my alma mater, 
the Free University of Brussels (Vrije Universiteit Brussel), and at Harvard University, has 
become very concerned about the disconcerting lack of concern about the possible dangers of 
AI. In an eye-opening opinion piece (Ginnis, 2025) he mentions that at a recent AI safety 
conference in Paris, no one seemed to be concerned any longer with the dangers of AI. Instead, 
it was all about PR and power struggles. Once, Ginnis states, AI safety was about risks for 
humanity. The possible threat that millions of people might lose their jobs, the danger that 
misinformation and manipulation would spread at a scale undermining democracy, the 
possibility that AI systems might one day take decisions we no longer comprehend, let alone 
control.  Instead, at the conference in Paris it was all about power. Who will get AI? Who 
 
 
 190 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
controls it? Who is running ahead in the race? The focus, according to Ginnis, shifted from risks 
to geopolitics. He issues a firm, unequivocal warning: 
“Humanity is creating a technology that surpasses its own knowledge and is 
completely unprepared for it. The first step is simple: acknowledge what is 
happening. The threat is real, the acceleration is dangerous, and the priorities 
are misplaced. There is still a long way to go, but we have to start 
somewhere.” (Ginnis, 2025, my translation) 
Likewise, Koen Schoors, professor of economy at the University of Ghent (Belgium), points to 
the danger of the growing attraction of an AI-based technocracy, not hampered by the 
sluggishness of democratic decision-making. There is no shortage of politicians, he states, who 
are fed up with the inertia of the democratic model with regard to the developments in AI 
(Schoors, 2024, p. 189).  
The question is not about whether or not or to what extent Harari, Hinton and others are right 
about their warnings. What intrigues me the most, as a philosopher of science, is: why are so 
many people so adamant and resolute in brushing aside all these warnings? Why do we cling so 
tightly to the reassuring idea that AI is just a mere tool, totally under our control? What are the 
tacit assumptions which apparently make it very hard for us to find the blind spots in our rosy 
vision on AI? 
2 Literature review: AI and philosophy of mind 
2.1 The Chinese room thought experiment 
When I was still a graduate student in philosophy, way back in the 1980s, I remember we were 
discussing John Searle’s thought experiment of the Chinese room. Remember that 40 years 
ago, AI was still a very remote theoretical possibility, a popular theme in science fiction movies 
perhaps, but not something to be taken very seriously. Nevertheless, some philosophers gave it 
their attention, more often than not in order to concoct sophisticated arguments to show that 
artificial intelligence would not be possible. A machine, a computer, could never have a mind in 
the same way human beings can be said to have minds.  
In brief, John Searle’s argument (Searle, 1980 and 1984) goes as follows. Someone is sitting in a 
room and receives a sheet of paper with Chinese characters from the left, through a slit in the 
wall. This person then meticulously follows a highly detailed instruction table, akin to a 
computer program, to transform the Chinese messages into other messages with different 
characters on another sheet of paper. Once the conversion is complete, this person then sends 
this new sheet of paper out to the right as output (likewise, through a slit in the wall). To an 
outside observer, it seems as if the person inside the rooms understands Chinese. However, in 
 
 
 191 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
reality, the person inside the room is just following instructions, he does not need to understand 
a single word of Chinese himself. 
Searle wanted to show with this thought experiment that two people can be functionally 
identical: a native Chinese speaker and the person inside the Chinese Room. They both provide 
perfect answers to questions posed in Chinese. Yet, they have completely different mental 
states (one understands Chinese, while the other does not understand it at all). The bottom line 
of Searle's argument is: a computer will remain fundamentally different from a human being. A 
computer system will never “truly” understand what it is doing, whereas human beings 
obviously can.  
2.2 Mary in the black-and-white room 
Another famous example in this respect comes from the Australian philosopher Frank Jackson, 
who in 1986 published a very controversial and influential article, What Mary didn’t know 
(Jackson, 1986). Even though it is almost forty years old by now, it is about a thought experiment 
which is still used today in discussions about the possibility of artificial general intelligence 
(AGI). 
In short, Jackson's thought experiment goes as follows. This experiment is about a fictional 
scientist in a distant future, Mary. In that distant future, both physics and neurophysiology have 
reached a final state. That means that Mary knows everything these sciences have to say about 
perceiving colors.  
But there is something special about Mary: she lives in a completely colorless room. Everything 
is black or white. So, Mary has never seen a color before in her entire life. Her knowledge of 
colors is therefore purely based on books about physiology, neurology, and the biochemistry of 
color perception. 
Now it gets exciting. One day, Mary finds a secret door to the outside world. The first thing she 
sees when she steps outside is a red apple. The question Jackson poses is: does Mary learn 
something new when she sees this apple?  
Jackson answers: yes, she learns a new fact — she learns what it is like to experience the color 
'red’. This phenomenon is referred to in philosophy of mind as ‘qualia’: ‘individual instances of 
subjective, conscious experience’ (qualia is Latin and is the plural of quale, which literally 
means ‘such as’). 
Jackson aims to conclude that knowledge about qualia fundamentally relies on subjective 
experiences, unlike knowledge about physical states in our brains. By subjective, we mean how 
someone experiences or judges something from a personal perspective.  
 
 
 192 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
Although Jackson’s knowledge argument was originally meant to question the philosophical 
position of physicalism, in short, the view that “everything is physical”, that there is nothing 
“above” the physical (van Biezen, 2016), this thought experiment has also become a classic in 
discussions about artificial general intelligence (Wang, 2023); Baron, 2025; Renard, 2024). 
What it boils down to is that Jackson’s argument about Mary in the black-and-white room is 
used to demonstrate that there will always be an unbridgeable gap between the human mind 
and artificial intelligence. In short, AI is Mary stuck in the black-and-white room, never having 
seen a color in her entire life. The human being is Mary when she walks into the outside world 
and sees a red apple. 
2.3 The Copernican trauma 
Somehow, I cannot shed the impression that these arguments resonate with older arguments 
pertaining to the difference between human beings and other animals. Time and again, 
throughout the ages, people have tried to pinpoint demarcation criteria to demonstrate that 
there is a fundamental difference in kind between human beings and animals, and certainly not 
a difference in degree. Part of the upheaval caused by the publication of Darwin’s On the Origin 
of Species (Darwin, 1859) in 1859 was precisely that: the horrifying idea that human beings are 
just another animal in the tree of evolution. It seems that the gist of these arguments resurfaces 
in the discussions about the nature of artificial intelligence. Benjamin Bratton, philosopher of 
technology and professor at the University of California, San Diego, sees AI as the next phase in 
the series of Copernican human decentering of the once thought privileged position of human 
beings in the center of the world, a “Copernican trauma” (Bratton, 2024): 
“What is today called ‘artificial intelligence’ reveals that intelligence, 
cognition and even mind (…) are not what they seem to be, not what they feel 
like and not unique to the human condition (…) intelligence itself is 
artificializable.” (Bratton, 2024)  
2.4 Roger Penrose on minds and machines 
One of the most famous defenders of the position that there is a fundamental chasm between 
the human mind and machines is the esteemed mathematician and mathematical physicist Sir 
Roger Penrose, who shared the 2020 Nobel Prize in physics for his ground-breaking work on 
black hole formation in the 1960s (which he accomplished together with Stephen Hawking who 
passed away in 2018, so, unfortunately, he could not be nominated anymore). As early as 1989, 
in his book The Emperor’s New Mind (Penrose, 1989), Penrose argued that human 
consciousness is essentially non-algorithmic, which implies that human consciousness in 
principle could never be embedded by what is called a conventional Turing machine, i.e. the 
theoretical model which underlies every possible computer.  
 
 
 193 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
Penrose’s argument is based on the so-called first incompleteness theorem by Kurt Gödel, 
undoubtedly one of the most important results in the foundations of mathematics of the 
twentieth century (for a comprehensive account, see Nagel & Newman,1958). Put very simply: 
this theorem gives a mathematical proof that any consistent and sufficiently rich axiomatic 
system of ordinary arithmetic with natural numbers (0, 1, 2, 3,…) and basic arithmetic functions 
like addition (+) and subtraction (-) will always contain true statements about natural numbers 
which cannot be proved nor disproved within that formal system itself. In other words, these 
statements are true, but they cannot be derived from the axioms (that is, by a step-by-step 
procedure, i.e. an algorithm).  
Penrose’s argument boils down to his claim that, contrary to that formal system, a sufficiently 
skilled mathematician (i.e. a human mind) is indeed capable of arriving at and formulating 
those true statements (which are unprovable and underivable from within that system). 
According to Penrose, Gödel's first incompleteness theorem tells us that no computer which 
works within a formal system F can prove the sentence  
G(F) = "This sentence cannot be proved in F."  
But, Penrose continues, we humans can just "see" the truth of G(F). Because if G(F) were false, 
then it would be provable, which leads to a paradox, an absurd result. So, the human mind is 
capable of doing something which not a single computer can do. Therefore, Penrose concludes, 
consciousness can't be reducible to computation. This implies that machines can never have 
human-like consciousness and, therefore, artificial intelligence would be forever out of human 
reach.  
Needless to say, there has been lot of discussion on Penrose’s argument, but that would be far 
beyond the scope of this article (for a brave attempt at refuting Penrose's use of Gödel's 
theorems, see Krajewski, 2015). The only reason why I bring it up here is the question: why is 
one of the smartest persons on this planet so persistent in trying to prove that the human mind 
will always “surpass” the capabilities of any possible machine? 
The tacit assumption underlying all these approaches seems to be that artificial general 
intelligence systems should possess consciousness in order to be called “truly” intelligent. All 
these thought experiments and arguments serve the same strategy: machines, artificial 
intelligence, cannot possibly attain consciousness or “self-awareness”, if you’d like, and, 
hence, machines, artificial intelligence, cannot possibly be called “really” intelligent. 
2.5 “Can machines think?” (Alan Turing) 
It is exactly this tacit assumption that intelligence somehow presupposes consciousness that I 
want to put into question. This is crucial to my argument, so I want to state it as explicitly as I 
can. For some reason or another, whatever it might be, some people still see computers as 
 
 
 194 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
inanimate, dumb machines which are incapable of thinking or feeling anything. And because 
computers cannot think or feel anything, they are not capable of making any decisions on their 
own. 
This leads us to one of the key questions in the field of philosophy of artificial intelligence: the 
fascinating cross-border field between philosophy of mind and the philosophy of computer 
science (Brey & Søraker, 2009) that explores the implications of artificial intelligence for getting 
grips on concepts like ‘intelligence’, ‘consciousness’, ‘free will’ etc. (for a comprehensive 
account of the current state of affairs in the philosophy of artificial intelligence, see Müller, 
2025). This key question is: can something like artificial intelligence exist at all? As Alan Turing, 
the father of the computer and of AI, put it succinctly: can machines think? (Turing, 1950, p. 
442). Turing himself found this question simply “too meaningless” to deserve discussion: 
“The original question, "Can machines think?" I believe to be too meaningless 
to deserve discussion. Nevertheless I believe that at the end of the century 
the use of words and general educated opinion will have altered so much that 
one will be able to speak of machines thinking without expecting to be 
contradicted” (Turing, 1950, p. 442) 
The interesting question is not ‘Can machines think?’, but ‘Can we let machines do things for 
which human beings would need intelligence?’. Saying that AI systems cannot “truly” think, that 
is to say, think as human beings, is the equivalent of saying that planes cannot “truly” fly 
because they can’t flap their wings. 
Admittedly, whether or not AI can really be called intelligent or not is still a controversial debate. 
Philosopher Luciano Floridi traces the issue down to a conceptual issue: either we enlarge our 
definition of what intelligence actually is so that it also includes artificial forms of it, or we widen 
our conception of agency so that it also encompasses artificial forms of agency which do not 
necessarily presuppose intelligence (Floridi, 2025). Floridi is in favor of the last option, viewing 
AI as “agency without intelligence” (Floridi, 2025). 
Others are more clearly opposed to AI agency and its implications, considering AI systems as 
intelligent. Even as early as 2009, scholar Joanna Bryson unequivocally warned against the 
danger of humanising robots and declared that “robots are fully owned by us” (Bryson, 2009, p. 
1). She argues: “In humanising them, we not only further dehumanise real people, but also 
encourage poor human decision making in the allocation of resources and responsibility” 
(Bryson, 2009, p.1). 
Why, you might ask, is that so important? Because this reassuring and soothing position that AI 
cannot possibly be called “really” or “truly” intelligent entails the risk of creating a dangerous 
blind spot. As long as we keep on seeing AI systems as a mere tool, as a “dumb” instrument, 
passively waiting for human instructions to do something, we put ourselves in danger by 
 
 
 195 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
overlooking the fact that these systems are getting more and more agency, that is to say, more 
and more autonomy and decision-making capabilities. If those rapidly increasing decision-
making capabilities of AI systems remain in our blind spot, we run a serious risk of continuing to 
outsource human decision-making to the point where it is no longer under our control. And this 
blind spot is related to the deeply ingrained tacit assumption that intelligence presupposes 
consciousness. As Yuval Harari strikingly noted, there is a widespread conviction that 
computers and AI systems simply are not capable of making decisions assumes that making 
decisions is predicated on having consciousness (Harari, 2024, p. 201). However, Harari 
continues, the fact that in human beings, as in other mammals, intelligence is often 
accompanied by consciousness does not allow us to extrapolate from humans and other 
mammals to all possible entities (Harari, 2024, p. 201).  
Admittedly, the issue of what 'consciousness' precisely is remains very slippery. Nevertheless, 
as scholars Patrick Krauss and Andreas Maier recently pointed out, most biologists nowadays 
consider consciousness as a gradual phenomenon which, in different levels of complexity, can 
also be found in animals (Krauss & Maier, 2025). In this respect, according to the integrated 
information theory (Giulio Tononi), the level of consciousness, the level of consciousness 
depends on the structure of the underlying substrate (i.e. the brain, for humans and other 
animals). The more coherent or connected a system is, the more conscious it is. In short, 
consciousness is related to the mutual interconnectedness of a system (Krauss & Maier, 2025). 
To cut all these ramifications short, let us simply erase ‘consciousness’ from the equation. The 
rapid evolution of AI compels us to radically rethink our understanding of concepts like 
‘intelligence’ and ‘consciousness’. As Blaise Agüera y Arcas and James Manyika put it 
succinctly: “We’re in paradigm-shifting territory” (Agüera y Arcas & Manyika, 2025). Meanwhile, 
we should focus on reducing our blind spot and getting past this pernicious stumbling block of 
keeping on seeing AI as a mere tool, a mere instrument which cannot operate without human 
supervision and which we have under our control.  
3 Methodology 
In this article, although I bring in different elements from various disciplines, the methodology 
followed is that of a philosophical argumentation, as the central thread of the questions 
examined are most closely related to philosophical issues in the field of philosophy of mind.  
4 Findings: The risk of our blind spot 
4.1 AI gaining in agency  
AI gaining in agency, becoming more and more autonomous and capable of independent 
decision-making, is not about Terminator-like robots rampaging through the streets in a killer 
 
 
 196 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
spree. As Mustafa Suleyman, the co-founder of DeepMind, one of the leading AI research 
laboratories, already pointed out in his book The Coming Wave (Suleyman & Bashkar, 2023):  
“Many technologies and systems are becoming so complex that they’re 
beyond the capacity of any one individual to truly understand them (… ) In AI, 
the neural networks moving toward autonomy are, at present, not 
explainable. You can’t walk someone through the decision-making process to 
explain precisely why an algorithm produced a specific prediction.” 
(Suleyman & Bashkar, 2023) 
When it comes to artificial general intelligence (AGI), the risk of the blind spot impeding us to 
see the real potential of danger becomes even greater. When bringing up the mere possibility of 
artificial general intelligence, you easily get your share of disbelief and laughter, conjuring up 
the famous image of the super-intelligent HAL 9000 computer on board of the spaceship U.S.S. 
Discovery in the motion picture 2001: A Space Oddysey (1968, Stanley Kubrick), refusing to 
open the air-lock, in that very calm, soothing tone: “I’m sorry Dave, I’m afraid I can’t do that”.  
By brushing aside the possibility of artificial general intelligence, we run the risk of turning a 
blind eye to the real dangers of the rapid development of current AI technology. The position we 
are in right now remembers me of the horrifying images of the 2004 tsunami in Sout-East Asia. 
Remember the footage of the receding water from the beaches, with people laughing at this 
strange phenomenon, small boats, sloops, suddenly lying dry at the sea floor, maybe some fish 
flopping in a remaining shallow pool, children still playing around. Even when they saw the wall 
of water looming in the distance, people’s warning systems still didn’t seem to be triggered, 
they curiously kept on staring in the distance, instead of deciding to run to higher grounds as 
quickly as possible. 
Lest I be dismissed as another doomsday prophet, I want to add a critical note to the rosy 
picture of AI being hailed as the next Industrial Revolution, lifting the future of the human 
workforce to a whole new level. A bit like in the lyrics of that song of Timbuk 3, “The future’s so 
bright, I gotta wear shades”. Admittedly, there are quite a number of economic and labor 
studies highlighting the risks of automation and AI for skills mismatch and workforce 
displacement. However, voices pleading for some caution are easily drowned out as too overly 
pessimistic. 
4.2 “AI is set to surpass us in speed and understanding” (Geoffrey Hinton) 
Moreover, there are some recent developments taking an ugly turn. At a recent conference, 
Geoffry Hinton made a very interesting point, which strengthens me in my argument that we are 
put on the  wrong track, set on the wrong foot, so to speak, when we continue to convince 
ourselves that AI systems are just dumb tools, not capable of “really” understanding what they 
are doing. Hinton’s point was that even with today’s large language models (LLMs), there is a 
 
 
 197 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
key difference between them and the way human memory works, and that is “AI’s unmatched 
ability to share knowledge” (Saso, 2025). Human beings pass information in small pieces, 
whereas AI has the ability to synchronize trillions of bits in the blink of an eye: 
“(…) It’s no competition,” he said. If intelligence is about learning and sharing 
knowledge, AI is set to surpass us in speed and understanding. It’s a “very 
scary conclusion,” Hinton said—a warning that highlights the need for 
consensus on AI’s capabilities. (…)” (Saso, 2025) 
In a recent paper in Science on how to manage extreme AI risks amid rapid technological 
progress, Hinton warns unequivocally:  
“(…) Increases in capabilities and autonomy may soon massively amplify AI’s 
impact, with risks that include large-scale social harms, malicious uses, and 
an irreversible loss of human control over autonomous AI systems. (…)” 
(Bengio, Hinton et al., 2024) 
Now, let us not get carried away by popular science fiction ideas of evil AI systems taking over 
the world. The danger of AI I want to talk to you about does not come from systems like HAL 
9000, the super-intelligent computer, eliminating all humans on board of the U.S.S. Discovery. 
The danger I want to shed a light on is coming from us. It is we who are putting ourselves in 
danger because we are turning a blind eye to the growing agency of AI systems. And we are 
doing so because we are held captive by an image, the image of an AI system as a mere tool, the 
iconic image of a computer as a simple box with a keyboard. Because we tacitly assume that 
agency presupposes consciousness, and since we are firmly convinced that AI systems cannot 
possibly acquire consciousness, they will not acquire autonomous agency, so, there is nothing 
to worry about, isn’t it? 
However, reality is catching up fast with us. Even now, as we speak, AI systems analyze 
tremendous amounts of data, they take decisions in fractions of a second, more and more 
without any human intervention. Just think about all the sophisticated algorithms which 
manage the content feed on social media. Think about the trading by algorithms on financial 
markets, medical diagnostics driven by AI systems. The point is: we rely more and more on AI 
systems to make decisions that once were the sole province of human beings.  
4.3 The need for vigilance 
As we are getting more and more comfortable towards these new technologies, as we are 
embracing them more deeply, we tend to become less vigilant over the technology, not to say 
downright lazy with regard to the use of all those systems. The principle of ‘least effort’ is a 
strong predictor of human behavior (Anderson & Rainie, 2023). It makes me think of my former 
life, when I was still working as a computer scientist for IT companies. In those days, you had to 
 
 
 198 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
be able to write sophisticated and elegant search queries in SQL-type languages to retrieve 
relevant information from a database. Nowadays, we just shout “Hey, Google!”, followed by a 
simplified query, often a very inelegant one, and we accept the result at face value, without 
giving it a second thought. The near infinite capacity of human beings to take things for granted 
will never fail to amaze me. Aren’t we lulling ourselves to sleep to quickly?  
A striking example of how this lazy attitude, blindly accepting what AI systems regurgitate, might 
lead us straight into disaster has been reported recently by Bastian Leibe, professor at the 
RWTH Aachen University in Germany (Leibe, 2025). When President Donald Trump from the 
United States announced his reciprocal tariffs on April 2nd, 2025, a number of people  noticed 
that these proposed tariffs are not related to the actual tariffs these countries charge on 
imports from the United States. Instead, they have been shown to correspond to the United 
States’ trade deficit divided by the United States import volume from that country, which, 
according to economists, does not make any sense at all from an economic perspective (Leibe, 
2025). 
Admittedly, at the time of writing of this article (early April 2025), president Trum had just 
announced his tariff plan. No scientific analyses or academic articles on this topic had been 
published as yet, as it was just discovered and signaled by a few academic scholars who are 
well versed in the field, like Bastian Leibe. Since April 2nd, a lot of economists have been trying to 
find out how on earth someone could come up with such an insane strategy. Until, Leibe 
continues his argument, someone found out that if you pose the question of tariff tables to 
current LLMs (like ChatGPT version 4o, Gemini 2.5pro et al.), they all propose tariffs which turn 
out to be very close to the tariffs in the list of president Trump. Leibe concludes that the most 
likely conclusion is that the Trump administration simply based its tariffs “on the unchecked 
outputs of an LLM” (Leibe, 2025): 
“This has real-world consequences. It is already sending economies into 
turmoil and it will cause worldwide harm and suffering. And it is sadly nothing 
that AI safety research could have prevented -- because the problem lay in 
front of the screen.” (Leibe, 2025) 
 
We see it happening all around us, as we speak. Society is becoming more and more complex. 
We are delegating human decision-making to sophisticated systems which involve storing our 
digital data and automating decision rules at an ever-increasing pace.  
In other words, we are lured into outsourcing our cherished decision-making and autonomy to 
AI systems step by step, and as these smart systems, powered by sophisticated machine-
learning, will rapidly augment their sophistication level in the next, say, ten years from now, we 
might lose the ability to keep on making decisions independently of these systems. Barry 
 
 
 199 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
Chudakov, founder and principal of Certain Research, sees the relationship between human 
beings and AI systems as  
“(…) a struggle between the determined fantasy of humans to resist (‘I’m 
independent and in charge and no, I won’t give up my agency!’) and the 
seductive power of technology designed to undermine that fantasy (‘I’m fast, 
convenient, entertaining! Pay attention to me!’)” (…)” (Anderson & Rainie, 
2023) 
5 Debate: Towards a Human–AI Symbiosis 
What is the way forward? Is the future so bleak as some of these findings tend to suggest? No, I 
think not. But it is time that we realize that we will have to face difficult questions, now that 
there is still time to do so. What are the things we, as human beings, really want agency over? 
What should we list as the conditions under which we will turn to AI to help us in making 
decisions? And, the most pernicious issue, under what conditions and tight control 
mechanisms are we prepared to outsource certain precisely defined decisions to AI systems? 
We do not have the luxury of turning a blind eye to these difficult questions. 
A very promising view on a possible way forward is offered by Somendra Narayan, a professor of 
Strategy and Innovation at the University of Amsterdam, The Netherlands. His idea is that if we 
want to come to grips with understanding the impact of AI on human agency, it is helpful to see 
this interaction in terms of a symbiosis (Narayan, 2024). Instead of looking at AI as “an external 
force eroding our autonomy”, it would be better to look at it in terms of “an augmentation of 
human capability, a co-evolution where humans and machines are learning from each other 
and influencing each other’s decision-making” (Narayan, 2024). Narayan admits: yes, AI 
presents certain risks for human agency, but also opportunities. In order to ensure that AI 
augments human autonomy instead of diminishing it, we must build systems “where 
transparency and ethical considerations are central to how AI operates” (Narayan, 2024). His 
core argument is: human agency does not exist in isolation. Human agency is part of a larger 
socio-technical system. Human decisions do not arise in isolation, they are shaped by both 
technological tools and societal structures. In his view, AI is simply the most recent layer in this 
system.  
Tyler Suard, an AI researcher and developer, formerly at Apple and Meta, places a critical note 
on this sunny idea of a fruitful AI-human combination. He gives several examples of 
experiments where AI turned out to perform better on its own than when working together in a 
combined human-AI team. He concludes: if AI can indeed outperform human-AI teams, 
massive job loss will be lurking around the corner (Tyler, 2024). He calls for a reality check: we 
need to prepare for that possibility where AI will be a powerful independent entity in the 
workforce (Tyler, 2024). He pleas for governments and organizations to start working on policies 
 
 
 200 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
and regulations that address potential job displacement. He thinks about “social safety nets, 
retraining programs, and incentives for industries that create new job opportunities” (Tyler, 
2024). 
I want to join this positive but cautious attitude towards the future of the human workforce 
(Boudry & Friederich, 2024). No, these developments certainly do not mean the end of human 
work. However, we must see to it that the transition to human-AI cooperation is not too 
disruptive, as this could be a threat to an inclusive and just division of labor, possibly even 
destabilizing society (van Biezen, 2024). In the past few years, the World Economic Forum 
stresses time and again in its annual jobs report that both analytical thinking and creative 
thinking are considered as the most wanted core skills in order to face the challenges of the 
oncoming transformation of the labor market (van Biezen, 2024, p. 58). In other words, 
companies focus on very high-level profiles with abstract cognitive skills as critical for that 
“brave new world” (Huxley, 1932). But what about the medium-level and low-level 
occupations? We need to be aware of the urgency of scaffolding social protection and support 
for those who run the risk of being forced out.  
The research group Inclusive Society (department Research & Expertise) at UCLL University of 
Applied Sciences (Leuven, Belgium) attaches a lot of importance to inclusion when it comes to 
the future of the human workforce. We focus on the question: how can we contribute to building 
a more inclusive and fairer world? From this perspective, we plea for the need of continuing 
policy debates on AI regulation taking explicitly into account the phenomenon of growing AI 
agency. 
6  Conclusions 
The study has shown that the prevailing assumption of artificial intelligence (AI) as a mere 
tool—an inert instrument incapable of independent reasoning or decision-making—no longer 
holds in light of current technological developments. Through a philosophical and conceptual 
analysis, it was demonstrated that AI systems are gaining increasing levels of autonomy and 
agency, allowing humans to outsource parts of their decision-making to machines. This 
process, if left unchecked, risks creating social and ethical blind spots with potentially 
destabilizing consequences for labour structures and democratic decision-making. At the same 
time, the transition toward human–AI cooperation presents opportunities for innovation, 
inclusion, and new forms of collaboration—provided it is governed by ethical oversight and 
clear regulatory frameworks. 
The paper contributes to the interdisciplinary dialogue between philosophy, technology studies, 
and the social sciences by reframing the debate on AI not in terms of technical functionality but 
in terms of agency. It advances a conceptual bridge between classical philosophical arguments 
about consciousness and contemporary issues of algorithmic autonomy, offering a framework 
 
 
 201 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
for understanding AI as an emergent actor within socio-technical systems. This philosophical 
contribution strengthens the conceptual foundations of ongoing debates on AI governance and 
the future of work. 
For management and organizations, the findings underline the importance of recognizing AI 
systems as active participants in decision-making processes rather than as passive tools. 
Managers and policymakers are encouraged to develop governance models that balance 
efficiency with ethical responsibility, ensuring human oversight, transparency, and 
accountability in the implementation of AI technologies. On a broader societal level, the paper 
highlights the need for continued public and policy debates on AI regulation, particularly 
regarding its implications for inclusion, social justice, and democratic control. 
The study is theoretical in nature and does not include empirical or quantitative data. Its 
findings are based on philosophical reasoning and conceptual synthesis, which limits its direct 
applicability to specific organizational contexts. The rapid evolution of AI technologies also 
means that some empirical examples may quickly become outdated. 
Future research should empirically investigate how AI agency manifests in real organizational 
and societal contexts. Comparative studies across industries or sectors could provide insight 
into how different forms of AI autonomy affect human decision-making, trust, and 
accountability. Interdisciplinary research combining philosophy, organizational studies, and AI 
ethics would help to refine theoretical models and translate them into practical guidelines for 
governance. 
 
References 
1. Agüerra y Arcas, B. & Manyika, J. (2025). AI Is Evolving – And Changing Our Understanding of 
Intelligence. In Noema, Berggruen Institute, 
2. https://www.noemamag.com/ai-is-evolving-and-changing-our-understanding-of-
intelligence/.  
3. Anderson, J. & Rainie, L. (2023). The Future of Human Agency. In Pew Research Center, 
https://www.pewresearch.org/internet/2023/02/24/the-future-of-human-agency/.  
4. Baron, S. (2025). Are a Machine's Thoughts Real? The Answer Matters Now More Than Ever. 
In Science Alert, https://www.sciencealert.com/are-a-machines-thoughts-real-the-answer-
matters-now-more-than-ever.  
5. Bengio, Y, Hinton, G. et al. (2024), Managing Extreme AI Risks Amid Rapid Progress, in 
Science, Vol. 384, Issue 6698, pp. 842-845.  
6. Boudry, M. & Friederich, S. (2024). The Selfish Machine. On the Power and Limitation of 
Natural Selection to Understand the Development of Advanced AI. In Philosophy of Science, 
PhilSci-Archive, preprint, https://philsci-archive.pitt.edu/23903/.  
 
 
 202 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
7. Boudry, M. (2025). The Selfish Machine. Will Humanity Be Subjugated by Superintelligent 
Ais?. In Maarten Boudry’s Substack, https://maartenboudry.substack.com/p/the-selfish-
machine.  
8. Bratton, B. (2024). The Five Stages of AI Grief. In Noema, Berggruen Institute, 
https://www.noemamag.com/the-five-stages-of-ai-grief/  
9. Brey, Ph. & Johnny H. Søraker, J. H. (2009). Philosophy of Computing and Information 
Technology. In Philosophy of Technology and Engineering Sciences, edited by Antonie 
Meijers, Amsterdam, Elsevier, pp. 1341–1407. 
10. Bryson, J. (2009). Robots Should Be Slaves. Published at Joanna Bryson Publications, 
https://www.joannajbryson.org/publications/robots-should-be-slaves-pdf.   
11. Darwin, C. (1859). On the Origin of Species by Means of Natural Selection or the 
Preservation of Favoured Races in the Struggle for Life. New York, D. Appleton and 
Company, 1861 (first edition 1859). 
12. Dhondt, S. & Dessers, E. (eds.)(2022). Robot zoekt collega. Uitgeverij Lannoo. [In Dutch; 
English title: Robot seeking colleague]. 
13. Douglas Heaven, W. (2023). Deep learning pioneer Geoffrey Hinton quits Google. In MIT 
Technology Review, 
https://web.archive.org/web/20230501125621/https://www.technologyreview.com/2023/
05/01/1072478/deep-learning-pioneer-geoffrey-hinton-quits-google/.  
14. Edwards, B. (2025). What does “PhD-level” AI mean? OpenAI’s rumored $20,000 agent plan 
explained. In Ars Technica, https://arstechnica.com/ai/2025/03/what-does-phd-level-ai-
mean-openais-rumored-20000-agent-plan-explained/.  
15. Ferguson, N. (2025). The Doom Nexus. In Niall Ferguson’s Time Machine, 
https://niallferguson.substack.com/p/the-doom-nexus.  
16. Floridi, L. (2025). AI as Agency without Intelligence: On Artificial Intelligence as a New Form 
of Artificial Agency and the Multiple Realisability of Agency Thesis. February 12, 2024. 
Available at http://dx.doi.org/10.2139/ssrn.5135645.  
17. Foreman, J. T. (2024). How to Make it as a Doomsday Prophet. In The Metaphor, 
https://www.taylorforeman.com/p/how-to-make-it-as-a-doomsday-prophet.  
18. Ginnis, V. (2025). Is er nog íémand bekommerd om de gevaren van AI? In De Standaard, 15 
February 2025, https://www.standaard.be/cnt/dmf20250214_96655287 [In Dutch; English 
title: Is there still anyone concerned about the dangers of AI?]. 
19. Harari, Y. N. (2024). Nexus. A Brief History of Information Networks from the Stone Age to AI. 
Vintage Publishing, Kindle Edition. 
20. Huxley, A. (1932). Brave New World. Pdf edition, Coradella Collegiate Bookshelf, 2004, 
http://collegebookshelf.net.  
21. Jackson, F. (1986). What Mary Didn’t Know. In The Journal of Philosophy, Vol. 83, No. 5 (May, 
1986), pp. 291-295. 
22. Kahneman, D. (2011). Thinking Fast and Slow. New York, Farrar, Straus and Giroux. 
 
 
 203 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
23. Krajewski, S. (2015). Penrose’s Metalogical Argument is Unsound. In Ladyman, J. et al. 
(eds.)(2015). Road to Reality with Roger Penrose. Kraków (Poland), Copernicus Center 
Press, p. 87-104. 
24. Krauss, P. & Maier A. (2025). De geest in de machine. In EOS Psyche & Brein, June 2025, pp. 
20-25 [In Dutch, English translation of the title: The Ghost in the Machine]. 
25. Ladyman, J. et al. (eds.)(2015). Road to Reality with Roger Penrose. Kraków (Poland), 
Copernicus Center Press. 
26. Leibe, B. (2025). Post on LinkedIn, 
https://www.linkedin.com/feed/update/urn:li:activity:7313873939691130880/.  
27. Lim, D. (2024). Why Yuval Noah Harari’s AI Doomsday Prophecies Are Misleading. In 
Medium, https://medium.com/@don-lim/why-yuval-noah-hararis-ai-doomsday-
prophecies-are-misleading-5541504ec3ab.  
28. Molek, N., Pulinx, R. & van Biezen, A. (eds.)(2024). Analysis of the State of the Art on the 
Future of Human Workforce. Scientific Report. Transform, European Union. 
29. Molek, N., van Biezen, A. & Velez, M. J. (2025), Book of Abstracts. International 
Interdisciplinary Conference Transform “The Future of Human Workforce”. Novo Mesto 
(Slovenia), FOS. 
30. Müller, V. (2025). Philosophy of AI. A Structured Overview. In Smuha, N. (ed.)(2025). The 
Cambridge Handbook of the Law, Ethics and Policy of Artificial Intelligence. Cambridge 
University Press, p. 40-58. 
31. Nagel, E. & Newman, J. R. (1958). Gödel’s Proof. New York, New York University Press. 
32. Narayan, S. (2024). AI and the Future of Human Agency: Are We Outsourcing Decision-
Making or Evolving with Machines?. In Medium, 
https://medium.com/@narayan.somendra/ai-and-the-future-of-human-agency-are-we-
outsourcing-decision-making-or-evolving-with-machines-78da6ba4475f . 
33. Newman, S. et al. (2019). AI & Agency. In 2019 Summer Institute on AI and Society, in AI 
Pulse, 26 September 2019, https://aipulse.org/ai-agency/?pdf=417.  
34. Palazzolo, S. and Weinberg, C. (2025). OpenAI Plots Charging $20,000 a Month For PhD-
Level Agents. In The Information, https://www.theinformation.com/articles/openai-plots-
charging-20-000-a-month-for-phd-level-agents.  
35. Pelley, S. (2024). "Godfather of Artificial Intelligence" Geoffrey Hinton on the promise, risks 
of advanced AI. In CBS News, https://www.cbsnews.com/news/geoffrey-hinton-ai-dangers-
60-minutes-transcript/.  
36. Penrose, R. (1989). The Emperor’s New Mind. Concerning Computers, Minds and The Laws 
of Physics. Oxford, Oxford University Press. 
37. Renard, V. et al. (2024). Mary Steps Out: Capturing Patient Experience through Qualitative 
and AI Methods. In NEJM AI, Vol. 1 No. 12, https://ai.nejm.org/doi/10.1056/AIp2400567.  
 
 
 204 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
38. Sapunov, G. (2023). Turing, “Intelligent Machinery. A Heretical Theory”, 1951. In Gonzo ML, 
https://gonzoml.substack.com/p/turing-intelligent-machinery-a-
heretical?utm_campaign=post&utm_medium=web.  
39. Saso, E. (2025). The path to safe, ethical AI: SRI highlights from the 2025 IASEAI conference 
in Paris. In Schwarz Reisman Institute for Technology and Society, University of Toronto. 
https://srinstitute.utoronto.ca/news/the-path-to-safe-ethical-ai.  
40. Satyanarayan, A. and Jones, G. M. (2024). Intelligence as Agency: Evaluating the Capacity of 
Generative AI to Empower or Constrain Human Action. In An MIT Exploration of Generative AI 
- From Novel Chemicals to Opera, https://mit-genai.pubpub.org/pub/94y6e0f8/release/2.  
41. Searle, J. (1980). Minds, Brains and Programs. In Behavioral and Brain Sciences, 3, pp. 417-
517. 
42. Searle, J. (1984). Minds, Brains and Science. Cambridge, Mass., Harvard university press. 
43. Schoors, K. (2024). Alles wordt anders. Gent, Borgerhoff & Lamberigts. [In Dutch; English 
title: Everything Will Be Different] 
44. Smuha, N. A. (ed.)(2025). The Cambridge Handbook of the Law, Ethics and Policy of 
Artificial Intelligence. Cambridge University Press. 
45. Suard, T. (2024). The Future of Work: AI May Not Need Us After All. In Medium, 
https://medium.com/@ceo_44783/the-future-of-work-ai-may-not-need-us-after-all-
5df8eae52ed9.  
46. Suleyman, M. & Bhaskar, M. (2023). The Coming Wave. Technology, Power, and the Twenty-
First Century’s Greatest Dilemma. New York, Crown. 
47. Turing, A. (1950). Computing Machinery and Intelligence. In Mind, 49, pp. 433-460. 
48. Turing, A. (1951). Intelligent Machinery. A Heretical Theory. https://gwern.net/doc/ai/1951-
turing.pdf.  
49. von Hoffman, C. (2025). Smarter AI means bigger risks – Why guardrails matter more than 
ever. In MarTech, https://martech.org/smarter-ai-means-bigger-risks-why-guardrails-
matter-more-than-ever/.   
50. van Biezen, A.F. (2016). A Case for Naturalism. In van Biezen, A.F., The Torch of 
Discovery, http://alexanderfvanbiezen.blogspot.com/2016/05/a-case-for-naturalism.html. 
51. van Biezen, A.F. (2022). Top-Down Cosmology and Model-Dependent Realism. A 
Philosophical Study of the Cosmology of Stephen Hawking and Thomas Hertog. Brussels, 
VUB Press. 
52. van Biezen, A. (2024). Emerging Skills for the Future Workforce. In Molek, N., Pulinx, R. and 
van Biezen, A. (eds.)(2024), Analysis of the State of the Art on the Future of Human 
Workforce. Scientific Report., Transform, European Union, p. 50-62. 
53. Van Biezen, A. (2025a). Abstract of ‘AI is Not a Tool’. In Molek, N., van Biezen, A. & Velez, M. 
J. (2025), Book of Abstracts. International Interdisciplinary Conference Transform “The 
Future of Human Workforce”. Novo Mesto (Slovenia), FOS, p. 8. 
 
 
 205 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
54. van Biezen, A.F. (2025b). AI is not just another tool. What keeps us in the blind spot?. In van 
Biezen, A.F., The Torch of Discovery, 
https://alexanderfvanbiezen.blogspot.com/2025/04/ai-is-not-just-another-tool.html.  
55. Verbinnen, L. (2025), AI-gebruik stijgt, maar ook onze bezorgdheid: ‘Techno-optimisme 
maakt plaats voor technorealisme’. In EOS Wetenschap. [In Dutch; English title: AI usage 
rises, but so does our concern: ‘Techno-optimism gives way to tech realism’.]  
https://www.eoswetenschap.eu/technologie/ai-gebruik-stijgt-maar-ook-onze-bezorgdheid-
techno-optimisme-maakt-plaats-
voor?utm_source=ActiveCampaign&utm_medium=mail&utm_campaign=eos_515.  
56. Walther C.C. (2025). Hybrid Intelligence: The Future of Human-AI Collaboration. In 
Psychology Today, https://www.psychologytoday.com/us/blog/harnessing-hybrid-
intelligence/202503/hybrid-intelligence-the-future-of-human-ai-collaboration. 
57. Wang, X. (2023). The Possibility of Artifical Qualia. In Communications in Humanities 
Research, https://doi.org/10.54254/2753-7064/6/20230083.  
58. Wiggers, K. (2025). OpenAI reportedly plans to charge up to $20,000 a month for specialized 
AI ‘agents’. In TechCrunch, https://techcrunch.com/2025/03/05/openai-reportedly-plans-
to-charge-up-to-20000-a-month-for-specialized-ai-agents/.  
 
 
 
 
*** 
Alexander van Biezen is a philosopher of science, graduated from the Free University of Brussels (Vrije Universiteit Brussel) 
as a Doctor of Philosophy and Moral Sciences. He specialized in the philosophy of cosmology, with a doctoral dissertation on 
the cosmological models of Stephen Hawking and Thomas Hertog. His book Top-Down Cosmology and Model-Dependent 
Realism (2022, VUB Press) is freely available online through the research portal of the Free University of Brussels 
(https://researchportal.vub.be/). He has an additional background in religious studies and in computer science. 
Currently, he is employed as a teacher of philosophy and religion with the Arcadia school group in Aarschot, Belgium. His 
main areas of interest are cosmology and the philosophy of artificial intelligence. 
Alexander van Biezen je filozof znanosti, doktoriral je na Svobodni univerzi v Bruslju (Vrije Universiteit Brussel) 
kot doktor filozofije in moralnih znanosti. Specializiral se je za filozofijo kozmologije, z doktorsko disertacijo o 
kozmoloških modelih Stephena Hawkinga in Thomasa Hertoga. Njegova knjiga Top-Down Cosmology and Model-
Dependent Realism (2022, VUB Press) je prosto dostopna na spletu preko raziskovalnega portala Svobodne 
univerze v Bruslju (https://researchportal.vub.be/). Ima tudi dodatno izobrazbo na področju religijskih študij in 
računalništva. Trenutno je zaposlen kot učitelj filozofije in religije v skupini šol Arcadia v Aarschotu v Belgiji. 
Njegova glavna področja zanimanja sta kozmologija in filozofija umetne inteligence. 
*** 
 
 
 
 
 
 
 
 
 
 
 206 
Izzivi prihodnosti / Challenges of the Future  Vol. 10 / no. 4 / November 2025 
Povzetek 
UI ni zgolj orodje 
Vpliv naraščajoče tvornosti umetne inteligence na prihodnost dela 
 
Raziskovalno vprašanje (RV): Katere so temeljne filozofske predpostavke, ki oblikujejo sodobno 
razumevanje umetne inteligence (UI) kot zgolj orodja, in kako te predpostavke vplivajo na naše 
dojemanje naraščajoče tvornosti UI ter njenega možnega vpliva na prihodnost dela? 
Namen: Članek kritično preučuje razširjeno domnevo, da sistemi UI ostajajo pasivna orodja, 
popolnoma pod nadzorom človeka. Raziskuje, kako nove oblike tvornosti UI – razumljene kot 
avtonomne oziroma polavtonomne sposobnosti odločanja – izpodbijajo to predstavo in kakšne 
posledice ima ta premik za človeško delo, etiko in družbeno stabilnost. 
Metoda: Raziskava uporablja filozofsko in konceptualno metodologijo, utemeljeno v filozofiji duha in 
filozofiji znanosti. Opira se na klasične miselne poskuse (Searlov “kitajski sobi”, Jacksonovo “Mary v 
črno-beli sobi” in Penroseove argumente o nealgoritmični zavesti) ter vključuje sodobne 
interdisciplinarne razprave o tvornosti, avtonomiji in zavesti UI. Analiza temelji na kritičnem pregledu 
literature, ki združuje filozofske, tehnološke in družbenopolitične vire. 
Rezultati: Ugotovitve kažejo, da predpostavka o UI kot »neum­nem orodju« ne vzdrži več. Dokazi o 
naraščajoči avtonomiji UI potrjujejo, da se postopki odločanja, ki so bili nekoč izključno v domeni 
človeka, vse pogosteje prenašajo na stroje. Takšno postopno prenašanje človeške tvornosti lahko 
povzroči družbene in etične slepe pege ter vodi do neenakih transformacij dela in izzivov upravljanja. 
Vendar pa lahko nadzorovan prehod k sodelovanju med človekom in UI spodbuja inovativnost in 
vključenost, če temelji na etičnem nadzoru in ustreznih regulativnih okvirih. 
Organizacija: Za organizacije raziskava poudarja potrebo po pravočasnem predvidevanju sprememb 
v delovnih strukturah in procesih odločanja, ki jih povzročajo sistemi UI z naraščajočo tvornostjo. 
Menedžerje in oblikovalce politik spodbuja k oblikovanju upravljavskih okvirov, ki ohranjajo človeški 
nadzor in hkrati omogočajo odgovorno sodelovanje z UI. 
Družba: Na družbeni ravni raziskava poudarja nujnost odprtega političnega in etičnega dialoga o 
regulaciji UI. Naslavljanje posledic avtonomije UI je ključno za ohranjanje človeške tvornosti, 
demokratične odgovornosti in družbene pravičnosti v digitalni dobi. . 
Originalnost: Članek prispeva k povezovanju filozofske refleksije in družbeno-tehnične analize, saj 
ponovno opredeljuje UI ne zgolj kot tehnološko orodje, temveč kot nastajajočega akterja v sistemih 
človeškega odločanja. Razvija koncept »tvornosti UI« kot osrednjo analitično perspektivo za 
razumevanje preobrazbe dela.  
Omejitve/nadaljnje raziskovanje: Raziskava je konceptualne narave in ne vključuje empiričnih 
podatkov. Nadaljnje raziskave bi morale empirično preučiti, kako organizacije in delavci v praksi 
doživljajo tvornost UI – na primer s pomočjo etnografskih ali organizacijskih študij primerov – ter 
raziskati politične in regulativne instrumente, ki bi lahko omilili tveganja, povezana z avtomatizacijo 
in tehnokratskim upravljanjem.  
 
Ključne besede: umetna inteligenca, tvornost UI, zavest UI, delovna sila, prihodnost dela, filozofija 
znanosti, filozofija duha.  
 
 
  This work is licensed under a Creative Commons Attribution 4.0 International License.   
 This journal is published by Faculty of Organisation Studies in Novo mesto. 
 
 
 
 
 207