Abstract
The rapid proliferation of artificial intelligence (AI) co-pilots across workplace settings has prompted urgent scholarly inquiry into their implications for workforce skills and labour market equity. This dissertation examines whether AI co-pilots narrow existing skills gaps or create new divides between augmented and non-augmented workers. Through a comprehensive synthesis of peer-reviewed literature, this study analyses the heterogeneous effects of AI assistance across different skill levels, occupations, and organisational contexts. The findings reveal a nuanced reality: AI co-pilots can compress performance differentials within specific roles by disproportionately benefiting lower performers, whilst simultaneously exacerbating structural inequalities at the labour market level between well-resourced, digitally skilled workers and those lacking access to AI tools or requisite training. Furthermore, evidence suggests that over-reliance on AI assistance may accelerate skill decay and impede intergenerational knowledge transfer. The dissertation concludes that the distributional consequences of AI co-pilots are contingent upon governance frameworks, access policies, and deliberate skill-development interventions, rather than being technologically predetermined. These findings carry significant implications for policymakers, human resource practitioners, and educational institutions navigating the evolving landscape of human-AI collaboration.
Introduction
The integration of artificial intelligence into workplace environments represents one of the most consequential technological transformations of the contemporary era. Unlike previous waves of automation that primarily affected routine manual tasks, the current generation of AI systems—frequently termed ‘co-pilots’—possesses capabilities that extend into cognitive, creative, and analytical domains previously considered exclusively human (Frank et al., 2019). These AI co-pilots, encompassing tools such as generative AI assistants, intelligent decision-support systems, and automated workflow enhancers, are fundamentally reshaping how workers perform their duties, acquire skills, and develop professional competencies.
The deployment of AI co-pilots raises profound questions regarding labour market equity and the distribution of technological benefits. Historically, technological innovations have exhibited paradoxical effects on workforce stratification, simultaneously creating opportunities for some whilst displacing others (Autor, 2015). The present moment demands rigorous examination of whether AI co-pilots will democratise access to high-level capabilities—effectively narrowing skills gaps by augmenting the abilities of less-skilled workers—or whether they will entrench and amplify existing inequalities by creating new divisions between those with access to AI augmentation and those without.
This inquiry carries substantial academic, social, and practical significance. Academically, it contributes to longstanding debates within labour economics and organisational studies concerning skill-biased technological change and the future of work. Socially, understanding the distributional consequences of AI deployment is essential for ensuring that technological progress translates into broadly shared prosperity rather than concentrated gains. Practically, organisations and policymakers require evidence-based guidance to design AI implementation strategies that promote workforce development and mitigate potential harms.
The urgency of this investigation is underscored by the accelerating pace of AI adoption across sectors. According to recent evidence, generative AI tools have achieved unprecedented rates of workplace integration, with significant proportions of knowledge workers now regularly utilising AI assistance for core professional tasks (Noy and Zhang, 2023). This rapid diffusion necessitates timely scholarly analysis to inform governance decisions before patterns of inequality become entrenched.
Aim and objectives
The overarching aim of this dissertation is to critically evaluate whether AI co-pilots narrow skills gaps within the workforce or create new divides between augmented and non-augmented workers.
To achieve this aim, the following specific objectives guide the investigation:
1. To synthesise existing empirical evidence on how AI co-pilots alter skill requirements and worker capabilities across different occupational contexts.
2. To examine the differential impacts of AI co-pilots on workers of varying skill levels, including analysis of within-role performance effects and broader labour market consequences.
3. To investigate the implications of AI co-pilot use for workplace learning, tacit knowledge development, and intergenerational skill transfer.
4. To identify the conditions and policy mechanisms under which AI co-pilots may either reduce or exacerbate workforce inequalities.
5. To develop evidence-based recommendations for organisations and policymakers seeking to maximise the equalising potential of AI co-pilots whilst mitigating risks of creating new workforce divides.
Methodology
This dissertation employs a literature synthesis methodology, systematically examining and integrating findings from peer-reviewed academic sources to address the research objectives. Literature synthesis represents an established approach within the social sciences for generating comprehensive understandings of complex phenomena by drawing upon multiple empirical studies and theoretical perspectives (Snyder, 2019).
The synthesis draws primarily upon the research corpus identified through systematic database searches, supplemented by targeted searches for additional high-quality sources from peer-reviewed journals, governmental publications, and recognised international organisations. Sources were selected based on relevance to the research questions, methodological rigour, and publication in reputable academic venues.
The analytical approach follows a thematic synthesis framework, wherein findings from individual studies are organised according to key conceptual dimensions: skill transformation effects, inequality mechanisms, within-role performance impacts, labour market stratification, and learning implications. This thematic organisation enables identification of convergent findings across diverse methodological approaches and sectoral contexts, whilst also highlighting areas of scholarly disagreement or insufficient evidence.
The synthesis prioritises recent scholarship to capture the rapidly evolving nature of AI technology and its workplace applications. Studies examining generative AI tools, intelligent assistants, and collaborative robotics were included, reflecting the broad conceptualisation of AI co-pilots as systems designed to augment rather than fully replace human workers.
Limitations of this methodological approach are acknowledged. Literature synthesis necessarily depends upon the quality and scope of available primary research. The relatively recent emergence of advanced AI co-pilots means that longitudinal evidence on their impacts remains limited. Additionally, publication bias may result in under-representation of null or negative findings regarding AI’s effects on workforce equity.
Literature review
The transformation of skill requirements in the age of AI
The deployment of AI systems across economic sectors has precipitated significant shifts in the skills that employers demand and workers must possess. A substantial body of evidence indicates that AI generally raises demand for digital, cognitive, and adaptive skills whilst simultaneously reducing demand for routine or mid-range skills (Cramarenco, Burcă-Voicu and Dabija, 2023; Morandini et al., 2023; Zirar, Ali and Islam, 2023). This pattern creates substantial pressure for upskilling and reskilling initiatives, as workers must develop new competencies to remain valuable in AI-augmented work environments.
Research conducted across multiple national contexts confirms the pervasiveness of these skill shifts. Shengelia (2025) documents significant employment structure changes attributable to AI adoption, with particular impacts on occupations characterised by routine cognitive tasks. Similarly, Babashahi et al. (2024) systematically review evidence on skill transformation in industrial settings, finding consistent patterns of increased demand for problem-solving, critical thinking, and technological fluency. Liang, Fan and Wang (2025) provide evidence from China demonstrating that AI-driven technological innovation correlates with employment transformation favouring workers with higher educational attainment and digital capabilities.
The nature of skill demands varies considerably by occupational category and organisational context. In professional and managerial roles, AI co-pilots tend to enhance rather than replace human judgment, creating demand for skills in AI oversight, interpretation of machine-generated outputs, and ethical evaluation of algorithmic recommendations (Einola and Khoreva, 2022). In contrast, roles characterised by routine information processing face more direct substitution pressures, reducing demand for the skills traditionally associated with such positions.
Augmentation versus automation paradigms
A crucial distinction within the literature concerns whether AI systems function primarily to augment human capabilities or to automate tasks previously performed by humans. This distinction carries profound implications for understanding skills gaps and workforce divides.
Evidence suggests that many AI implementations augment workers’ capabilities through task support, enhanced decision-making, and accelerated learning rather than fully replacing human workers, particularly in high- and mid-skill roles (Holm and Lorenz, 2021; Frank et al., 2019; Zirar, Ali and Islam, 2023). Harborth and Kümpers (2021) conceptualise this phenomenon as ‘intelligence augmentation’, wherein AI systems amplify human cognitive abilities and enable workers to perform at higher levels than would otherwise be possible. De Assis Dornelles, Ayala and Frank (2023) examine collaborative robotics in manufacturing contexts, finding that properly designed human-robot collaboration can enhance worker skills rather than merely substituting for them.
However, the augmentation-automation distinction is not absolute and varies by deployment context. Marguerit (2025) provides compelling evidence that augmentation-oriented AI raises wages and generates new work primarily for high-skilled workers, whilst automation-oriented AI disproportionately harms low-skilled jobs and wages. This differential pattern suggests that even nominally augmentative AI may exacerbate labour market stratification when its benefits accrue predominantly to already-advantaged workers.
Within-role performance effects
A particularly significant strand of research examines how AI co-pilots affect performance differentials among workers occupying similar roles. This within-role perspective offers crucial insights into whether AI tools compress or amplify skill-based inequalities at the individual level.
Noy and Zhang (2023) conducted experimental research examining productivity effects of generative AI, finding that ChatGPT-style tools boosted productivity most substantially for lower performers on writing tasks, effectively compressing performance gaps within occupational categories. This finding suggests that AI co-pilots may function as equalising tools within specific work contexts by disproportionately benefiting those with initially weaker capabilities.
Ide (2025) corroborates and extends these findings, demonstrating that AI assistance can offset learning deficits and enable workers with less experience or training to perform comparably to more skilled colleagues. This performance-compressing effect represents a potentially significant mechanism through which AI co-pilots might narrow skills gaps within organisations.
Callari and Puppione (2025) provide qualitative evidence from a multinational corporation examining employee perceptions of generative AI productivity tools. Their findings indicate that workers perceive AI assistance as supportive of workplace learning, enriching knowledge, and enabling informal on-the-job development and collaboration. These perceived benefits appear particularly pronounced among workers who previously lacked access to developmental resources or mentorship.
Labour market stratification effects
Whilst evidence on within-role effects suggests skill-gap-narrowing potential, research examining broader labour market dynamics presents a more concerning picture of AI’s distributional consequences.
At the labour market level, benefits of AI co-pilot adoption concentrate where digital skills and training are already strong, widening divides between high-skill, AI-using workers and others (Holm and Lorenz, 2021; Frank et al., 2019; Zirar, Ali and Islam, 2023; Shengelia, 2025). This stratification pattern reflects multiple reinforcing mechanisms: workers with stronger baseline digital skills are more likely to have access to AI tools, better able to utilise them effectively, and positioned in occupations where AI augmentation generates the greatest productivity gains.
Georgieff and Hyee (2022) provide cross-country evidence demonstrating substantial variation in AI’s employment effects based on national contexts, with implications for inequality depending upon existing skill distributions and institutional frameworks. Their analysis suggests that countries with more unequal skill distributions and weaker training infrastructures may experience more pronounced labour market polarisation from AI adoption.
Marguerit (2025) distinguishes between AI’s effects on new work creation versus existing employment, finding that augmentation AI generates new work opportunities predominantly for high-skilled workers whilst providing limited benefits to those in lower-skill positions. This pattern suggests that even when AI does not directly displace workers, its benefits may flow disproportionately to already-advantaged labour market participants.
Workplace learning and skill development implications
The relationship between AI co-pilot use and skill development represents a critical but relatively underexplored dimension of the literature. Evidence here is mixed, suggesting both potential benefits and significant risks.
On the positive side, generative AI assistants and co-pilots can support workplace learning by enriching knowledge resources and enabling informal, on-the-job development and collaboration (Callari and Puppione, 2025; Wilkens, 2020). Workers may utilise AI assistance to explore unfamiliar domains, receive immediate feedback on work products, and access expertise that would otherwise require extensive training or experienced mentors.
However, concerning evidence has emerged regarding AI’s potential to accelerate skill decay and undermine incentives for deep learning. Macnamara et al. (2024) examine whether using AI assistance accelerates skill decay and hinders skill development, finding evidence that performers may not be aware of their declining independent capabilities when habitually relying upon AI support. This finding raises significant concerns about the long-term consequences of AI co-pilot dependence.
Ide (2025) extends this analysis by examining intergenerational knowledge transmission, suggesting that AI co-pilots may disrupt traditional mechanisms through which experienced workers transfer tacit knowledge to newer colleagues. When novice workers can achieve adequate performance through AI assistance, incentives for developing deep expertise through mentored practice may diminish, potentially creating hidden deficits in future workforce capabilities.
Holm and Lorenz (2021) document heterogeneous within-job effects, finding that whilst AI can upgrade skill requirements in some occupations, it may simultaneously increase work pace and reduce autonomy in others, with implications for workers’ opportunities to engage in developmental activities.
Governance and access as mediating factors
A crucial finding emerging across the literature concerns the role of governance frameworks, access policies, and deliberate skill-development interventions in shaping AI’s distributional consequences. These factors appear to largely determine whether skill-gap-narrowing or divide-widening effects predominate in any given context.
Access to AI co-pilots combined with appropriate training can lift lower- or mid-performers within given roles, narrowing internal gaps (Callari and Puppione, 2025; Ide, 2025; Noy and Zhang, 2023). However, realising this potential requires deliberate organisational investment in ensuring broad access and supporting workers in developing effective AI utilisation capabilities.
Zirar, Ali and Islam (2023) emphasise the importance of thoughtful approaches to human-AI coexistence, identifying organisational factors that influence whether AI deployment supports or undermines worker development. Their research agenda highlights the need for further investigation into governance mechanisms that can maximise AI’s beneficial effects whilst mitigating risks.
Frank et al. (2019) argue that policy interventions play essential roles in determining how AI’s labour market effects unfold, suggesting that technological outcomes are substantially shaped by institutional choices rather than being technologically predetermined. This perspective underscores the agency that policymakers and organisations possess in shaping AI’s consequences for workforce equity.
Discussion
Reconciling apparently contradictory evidence
The evidence synthesised in this dissertation presents an apparent paradox: AI co-pilots simultaneously narrow performance gaps within specific roles whilst widening structural divides across the broader labour market. Resolving this apparent contradiction requires attention to the multiple levels at which AI’s effects operate.
At the individual task level, AI co-pilots function as capability amplifiers that disproportionately benefit workers with initially weaker skills or performance. The experimental evidence from Noy and Zhang (2023) demonstrates this pattern clearly: when all workers in a given role receive access to AI assistance, performance gaps compress because lower performers experience larger proportional gains. This mechanism represents a genuine skill-gap-narrowing effect operating within defined occupational boundaries.
However, this within-role equalisation occurs against a backdrop of labour market stratification wherein access to AI co-pilots is not uniformly distributed. Workers in high-skill occupations, organisations with substantial technological resources, and sectors at the forefront of AI adoption gain access to augmentation tools whilst others do not. The result is a widening divide between AI-augmented and non-augmented workers, even as gaps narrow among those with AI access.
Furthermore, the skill-gap-narrowing effects observed in controlled experimental settings may not fully manifest in real-world deployment contexts. Organisational factors including training provision, workplace culture, and managerial support influence whether workers can effectively leverage AI tools. Where these supportive conditions are absent, providing AI access alone may prove insufficient to realise equalising potential.
The skill decay dilemma
The evidence regarding AI’s effects on skill development presents particularly challenging implications for policymakers and organisations. Whilst AI co-pilots can support immediate performance and even facilitate certain forms of workplace learning, they may simultaneously undermine the deep practice and experiential learning through which durable expertise develops.
This skill decay dynamic represents a potential hidden cost of AI augmentation that may not become apparent until substantial time has elapsed. Workers who achieve satisfactory performance through AI assistance may not recognise their declining independent capabilities until circumstances require unaided performance. The evidence from Macnamara et al. (2024) suggests that this awareness gap is particularly concerning, as workers may overestimate their retained skills.
The implications extend beyond individual workers to organisational and societal levels. If AI co-pilots disrupt intergenerational knowledge transmission, as Ide (2025) suggests, organisations may find themselves with diminished pools of deep expertise from which to draw for complex problem-solving, innovation, and training of future workers. This represents a form of human capital depreciation that may be difficult to reverse once established.
Addressing the skill decay dilemma requires deliberate design of AI co-pilot implementations that preserve opportunities for unassisted practice and maintain strong incentives for developing foundational competencies. This may involve periodic ‘AI-free’ work periods, assessment of independent capabilities alongside AI-augmented performance, and explicit attention to mentorship and tacit knowledge transfer in workforce development strategies.
Implications for the augmented versus non-augmented divide
The central question of whether AI co-pilots create a new divide between augmented and non-augmented workers admits no simple answer based upon current evidence. The nature and magnitude of such a divide depends substantially upon contextual factors including technological access, training provision, and governance frameworks.
Under conditions of broad access and strong skill-development support, AI co-pilots may function as democratising tools that enable workers across skill levels to perform at higher levels. The within-role performance compression documented by Noy and Zhang (2023) and related studies demonstrates the potential for AI to narrow gaps when access is equalised.
However, under conditions of uneven access and inadequate training support, AI co-pilots risk creating new forms of labour market segmentation. Workers with AI augmentation may enjoy substantial productivity advantages, enhanced earnings, and expanded opportunities, whilst non-augmented workers face relative decline in their labour market positions. The evidence from Marguerit (2025) regarding differential effects on high-skilled versus low-skilled workers suggests that this stratification pattern is already emerging.
The divide between augmented and non-augmented workers may prove particularly pernicious because it compounds existing inequalities. Workers with higher initial skill levels and greater organisational resources are more likely to gain early access to AI co-pilots, enabling them to capture disproportionate benefits and further extend their advantages over less-resourced counterparts.
Meeting research objectives
The analysis undertaken in this dissertation has addressed each of the stated research objectives. Objective one, concerning skill requirement changes, has been addressed through synthesis of evidence demonstrating that AI co-pilots generally elevate demand for digital, cognitive, and adaptive skills whilst reducing demand for routine capabilities, creating substantial upskilling and reskilling pressures.
Objective two, examining differential impacts by skill level, has been addressed through analysis of both within-role effects (where AI tends to benefit lower performers proportionally more) and labour market effects (where benefits concentrate among already high-skilled workers). This two-level analysis reveals the apparent paradox at the heart of AI’s distributional consequences.
Objective three, concerning workplace learning implications, has been addressed through examination of evidence on both learning-supportive and learning-undermining effects of AI co-pilots. The skill decay dilemma and concerns regarding intergenerational knowledge transfer emerge as significant considerations requiring attention in AI implementation strategies.
Objective four, identifying conditions shaping AI’s distributional effects, has been addressed through analysis of governance, access, and training as key mediating factors. The evidence consistently indicates that these contextual elements substantially determine whether skill-gap-narrowing or divide-widening effects predominate.
Objective five, concerning policy and practice recommendations, is addressed in the concluding section based upon the preceding analysis.
Conclusions
This dissertation has examined whether AI co-pilots narrow skills gaps or create new divides between augmented and non-augmented workers. The analysis reveals that AI co-pilots are not inherently equalising or polarising; rather, their distributional consequences depend substantially upon deployment contexts, access policies, and deliberate skill-development interventions.
Within roles where workers have equalised access to AI co-pilot tools, evidence consistently demonstrates performance gap compression, with lower performers experiencing the largest proportional benefits from AI assistance. This within-role equalisation represents a genuine skill-gap-narrowing mechanism that organisations can leverage through inclusive AI deployment strategies combined with appropriate training and support.
At the broader labour market level, however, AI co-pilots risk widening structural divides between well-trained, AI-augmented workers and those in low-skill or non-augmented positions. Benefits concentrate where digital skills and technological resources are already strong, potentially exacerbating existing patterns of labour market inequality. This stratification effect represents a significant policy concern requiring proactive intervention.
Furthermore, the evidence regarding skill decay and disrupted intergenerational knowledge transfer suggests that poorly designed reliance on AI assistants can erode tacit expertise and create hidden deficits for future workers. Organisations must balance the immediate productivity benefits of AI augmentation against longer-term human capital development considerations.
Based upon this analysis, several recommendations emerge. Organisations should prioritise broad access to AI co-pilot tools rather than limiting deployment to already high-performing workers or elite occupational categories. Training investments should accompany technology deployment to ensure workers can effectively leverage AI assistance. Assessment frameworks should evaluate both AI-augmented and independent performance to maintain incentives for developing foundational skills. Mentorship and experiential learning opportunities should be preserved to support intergenerational knowledge transfer.
For policymakers, the evidence underscores the importance of skills development infrastructure investment, digital literacy initiatives, and regulatory frameworks that promote equitable access to AI capabilities. Left unaddressed, market forces may produce AI deployment patterns that exacerbate labour market polarisation.
Future research should examine the long-term consequences of AI co-pilot use for skill development and workforce capabilities through longitudinal study designs. Investigation of effective governance mechanisms for promoting equitable AI deployment would inform policy development. Comparative research across national and institutional contexts would illuminate how different regulatory and educational frameworks shape AI’s labour market effects.
In conclusion, AI co-pilots present both significant opportunities for narrowing skills gaps and substantial risks of creating new workforce divides. The balance between these outcomes is not technologically predetermined but shaped by human choices regarding governance, access, and skill development. Evidence-based policy and practice can maximise the equalising potential of AI augmentation whilst mitigating the risks of deepening labour market inequalities.
References
Autor, D., 2015. Why are there still so many jobs? The history and future of workplace automation. *Journal of Economic Perspectives*, 29(3), pp. 3-30.
Babashahi, L., Barbosa, C., Lima, Y., Lyra, A., Salazar, H., Argôlo, M., Almeida, M. and Souza, J., 2024. AI in the workplace: a systematic review of skill transformation in the industry. *Administrative Sciences*, 14(6), 127. https://doi.org/10.3390/admsci14060127
Callari, T. and Puppione, L., 2025. Can generative artificial intelligence productivity tools support workplace learning? A qualitative study on employee perceptions in a multinational corporation. *Journal of Workplace Learning*. https://doi.org/10.1108/jwl-11-2024-0258
Cramarenco, R., Burcă-Voicu, M. and Dabija, D., 2023. The impact of artificial intelligence (AI) on employees’ skills and well-being in global labor markets: a systematic review. *Oeconomia Copernicana*, 14(3), pp. 731-767. https://doi.org/10.24136/oc.2023.022
De Assis Dornelles, J., Ayala, N. and Frank, A., 2023. Collaborative or substitutive robots? Effects on workers’ skills in manufacturing activities. *International Journal of Production Research*, 61(23), pp. 7922-7955. https://doi.org/10.1080/00207543.2023.2240912
Einola, K. and Khoreva, V., 2022. Best friend or broken tool? Exploring the co-existence of humans and artificial intelligence in the workplace ecosystem. *Human Resource Management*, 62(1), pp. 117-135. https://doi.org/10.1002/hrm.22147
Frank, M., Autor, D., Bessen, J., Brynjolfsson, E., Cebrián, M., Deming, D., Feldman, M., Groh, M., Lobo, J., Moro, E., Wang, D., Youn, H. and Rahwan, I., 2019. Toward understanding the impact of artificial intelligence on labor. *Proceedings of the National Academy of Sciences of the United States of America*, 116(14), pp. 6531-6539. https://doi.org/10.1073/pnas.1900949116
Georgieff, A. and Hyee, R., 2022. Artificial intelligence and employment: new cross-country evidence. *Frontiers in Artificial Intelligence*, 5, 832736. https://doi.org/10.3389/frai.2022.832736
Harborth, D. and Kümpers, K., 2021. Intelligence augmentation: rethinking the future of work by leveraging human performance and abilities. *Virtual Reality*, 26, pp. 849-870. https://doi.org/10.1007/s10055-021-00590-7
Holm, J. and Lorenz, E., 2021. The impact of artificial intelligence on skills at work in Denmark. *SSRN Electronic Journal*. https://doi.org/10.2139/ssrn.3876347
Ide, E., 2025. Automation, AI, and the intergenerational transmission of knowledge. Working paper.
Liang, H., Fan, J. and Wang, Y., 2025. Artificial intelligence, technological innovation, and employment transformation for sustainable development: evidence from China. *Sustainability*, 17(9), 3842. https://doi.org/10.3390/su17093842
Macnamara, B., Berber, I., Çavusoglu, M., Krupinski, E., Nallapareddy, N., Nelson, N., Smith, P., Wilson-Delfosse, A. and Ray, S., 2024. Does using artificial intelligence assistance accelerate skill decay and hinder skill development without performers’ awareness? *Cognitive Research: Principles and Implications*, 9, 72. https://doi.org/10.1186/s41235-024-00572-8
Marguerit, D., 2025. Augmenting or automating labor? The effect of AI development on new work, employment, and wages. *SSRN Electronic Journal*. https://doi.org/10.2139/ssrn.5169611
Morandini, S., Fraboni, F., Angelis, M., Puzzo, G., Giusino, D. and Pietrantoni, L., 2023. The impact of artificial intelligence on workers’ skills: upskilling and reskilling in organisations. *Informing Science: The International Journal of an Emerging Transdiscipline*, 26, pp. 39-68. https://doi.org/10.28945/5078
Noy, S. and Zhang, W., 2023. Experimental evidence on the productivity effects of generative artificial intelligence. *Science*, 381(6654), pp. 187-192. https://doi.org/10.1126/science.adh2586
Shengelia, R., 2025. Artificial intelligence and labor market dynamics: employment problems and development trends. *Economics*, 107(3-5), pp. 7-15. https://doi.org/10.36962/ecs107/3-5/2025-07
Snyder, H., 2019. Literature review as a research methodology: an overview and guidelines. *Journal of Business Research*, 104, pp. 333-339.
Wilkens, U., 2020. Artificial intelligence in the workplace – a double-edged sword. *The International Journal of Information and Learning Technology*, 37(5), pp. 253-265. https://doi.org/10.1108/ijilt-02-2020-0022
Zirar, A., Ali, S. and Islam, N., 2023. Worker and workplace artificial intelligence (AI) coexistence: emerging themes and research agenda. *Technovation*, 124, 102747. https://doi.org/10.1016/j.technovation.2023.102747
