Nearly a century ago, economist John Maynard Keynes predicted that by the early 21st century, technological advancements would reduce the workweek to just 15 h. However, as we now recognize, this prediction has not been realized. Kaplan et al. (Reference Kaplan, Aitken, Allan, Alliger, Ballard and Zacher2025) offer insightful explanations for why Keynes’ forecast has not materialized.
Yet, as we stand on the brink of a new technological revolution—the era of generative artificial intelligence (AI)—we must ask ourselves: Could Keynes’ vision of reduced work hours finally become a reality? This paper seeks to answer this question by examining the potential impact of generative AI on work hours and productivity through the lens of different worker types and their adaptability to AI technologies.
Recent research has demonstrated the transformative potential of generative AI in the workplace. Brynjolfsson et al. (Reference Brynjolfsson, Li and Raymond2023) found that customer support agents using AI resolved 14% more issues per hour, with the most substantial benefits seen among novice and low-skilled workers. In another study, professionals using ChatGPT for writing tasks completed their work 40% faster. These findings suggest that generative AI could indeed lead to a reduction in work hours, as tasks are completed more quickly and efficiently (Noy et al., Reference Noy and Zhang2023).
However, we propose that the impact of generative AI on work hours and productivity will not be uniform across all workers. To better understand these differential impacts, we present a framework based on two key dimensions: AI adaptability and work ethic. This framework draws inspiration from Everett M. Rogers’ diffusion of innovation theory (Rogers, Reference Rogers1962), adapting it to the individual level in the context of AI adoption in the workplace. Whereas Rogers’ theory focuses on the spread of innovations through a social system, our framework examines how individuals within an organization might adopt and adapt to AI technologies. This framework aligns with key constructs in I-O psychology: adaptability, which relates to the broader concept of individual adaptability in the workplace (Pulakos et al., Reference Pulakos, Arad, Donovan and Plamondon2000), and work ethic, which can be linked to the conscientiousness factor in the five-factor model of personality (Costa & McCrae, Reference Costa and McCrae1992).
Our choice to focus on AI adaptability and work ethic as key dimensions for our typology stems from their crucial role in determining how individuals interact with and benefit from AI technologies in the workplace. AI adaptability refers to an individual’s capacity and willingness to effectively learn, integrate, and leverage AI technologies in their work processes. If we consider its personality equivalent, AI adaptability would closely relate to openness to experience from the five-factor model of personality. This connection reflects an individual’s propensity to embrace novel ideas and technologies, and their willingness to integrate them into their workflows. Equally important is work ethic, which aligns closely with the conscientiousness factor. Work ethic influences how committed workers remain to diligent effort, resisting the temptation to rely on easy but potentially ineffective shortcuts that AI might offer. By combining these two dimensions, we create a comprehensive framework that captures both the willingness to adopt new technologies and the drive to use them effectively and responsibly.
It is important to note that these dimensions represent a continuum rather than discrete categories. Workers may fall anywhere along these spectrums, and their positions may shift over time. However, for the sake of clarity in our analysis, we focus on the four archetypal categories that emerge from the extremes of these dimensions.
Worker types in the AI era
Our framework identifies four distinct worker types: laggards, diligent conservatives, surface adopters, and AI augmenters. We will discuss each of these types in turn, considering implications for work hours, productivity, and worker well-being.
Laggards
Laggards, characterized by low work ethic and low AI adaptability, face the greatest challenges in the AI-driven workplace. These individuals may score low on both conscientiousness and openness to experience, exhibiting low levels of intrinsic motivation (Ryan & Deci, Reference Ryan and Deci2000) and resistance to organizational change (Oreg, Reference Oreg2003). In the short term, laggards may see little change in their work hours or productivity, as they neither put in extra effort nor adopt new technologies. However, in the long term, they are likely to experience involuntary decreases in work hours due to reduced demand for their skills. Their productivity is expected to decline significantly relative to the workers who embrace AI, potentially leading to job displacement. Interestingly, recent research suggests that generative AI can help reduce productivity inequality among workers by providing more support to those with weaker skills (Noy et al., Reference Noy and Zhang2023). This could potentially benefit laggards if they can be encouraged to adopt these tools. From an I-O psychology perspective, interventions for this group might need to focus on increasing work engagement and fostering a growth mindset.
In terms of well-being, laggards may initially experience less stress related to technological change because they avoid the learning curve associated with adopting new AI technologies. By resisting or delaying the integration of AI into their work processes, they temporarily sidestep the cognitive demands and potential frustrations that come with learning new systems and methods. However, the lower levels of stress are likely to be short-lived. As AI technologies become more prevalent and integral to work processes, laggards may find themselves increasingly out of step with workplace expectations and requirements, potentially resulting in job insecurity, decreased job satisfaction, and other negative impacts on their long-term well-being.
Diligent conservatives
Diligent conservatives, on the other hand, are characterized by their strong work ethic and dedication to their craft, but they struggle with adopting new technologies. These individuals likely score high on conscientiousness but low on openness to experience (Costa & McCrae, Reference Costa and McCrae1992). Their resistance to technological change may stem from various factors, including low self-efficacy for using AI tools or a belief in the superiority of traditional methods. In the short term, diligent conservatives may maintain their productivity through sheer effort and expertise in their field. They might even see a temporary increase in work hours as they attempt to compete with AI-augmented colleagues. However, this approach is likely unsustainable. In the long term, their productivity may stagnate or decline relative to AI-adopting peers, potentially leading to increased stress and burnout. To support diligent conservatives, organizations would need to focus on targeted training programs that address the psychological barriers to technology adoption. This could include interventions aimed at increasing technology acceptance and building AI self-efficacy.
The well-being of diligent conservatives may suffer in the short term due to increased stress and potential burnout as they work harder to compete with AI-augmented colleagues. However, their strong work ethic may provide a sense of purpose and accomplishment. In the long term, if they can adapt to AI technologies, they may experience improved well-being through reduced workload and increased efficiency.
Surface adopters
Surface adopters present a unique case of individuals who readily adopt new AI technologies but lack the underlying work ethic or deep skills to fully leverage these tools. They likely score high on openness to experience but low on conscientiousness. Although they demonstrate high computer self-efficacy, they may lack other important work-related competencies. In the short term, surface adopters might see a decrease in work hours and an increase in productivity as they quickly adopt AI tools to automate aspects of their work. This aligns with the observed productivity boosts in recent studies (Brynjolfsson et al., Reference Brynjolfsson, Li and Raymond2023). However, this initial boost may be misleading. In the long term, their lack of deep skills and work ethic could become apparent, potentially leading to stagnation in productivity. They may find themselves needing to increase work hours to compensate for their limitations, especially when competing with more diligent AI users. The case of surface adopters highlights the importance of balancing technological skills with core work competencies.
Surface adopters might initially experience improved well-being due to reduced work hours and increased productivity. However, this could be short lived. In the long term, their lack of deep skills may lead to job insecurity and stress, potentially negatively impacting their well-being. They may also miss out on the sense of accomplishment that comes from mastering complex skills.
AI augmenters
Finally, AI augmenters represent the ideal adaptation to the AI revolution. These individuals combine a strong work ethic with high AI adaptability, likely scoring high on both conscientiousness and openness to experience. They also demonstrate high levels of learning agility (De Meuse et al., Reference De Meuse, Dai and Hallenbeck2010), quickly acquiring and applying new skills in AI integration. In the short term, AI augmenters may maintain or even increase their work hours as they invest time in learning and integrating AI tools into their workflows. However, their productivity is likely to see a significant boost, even more than observed in the recent research findings (Brynjolfsson et al., Reference Brynjolfsson, Li and Raymond2023; Noy et al., Reference Noy and Zhang2023). In the long term, these workers have the potential to realize Keynes’ vision of reduced work hours coupled with high productivity. They may be able to accomplish more in less time, potentially leading to a reduction in overall work hours without sacrificing output. However, their higher achievement may also prompt them to strive for greater success (Hirschi & Spurk, Reference Hirschi and Spurk2021), ultimately leading to no reduction in their work hours. The success of AI augmenters raises important questions about job design and work distribution. Organizations may need to reconsider how they structure roles and responsibilities to fully leverage the capabilities of these highly productive individuals.
AI augmenters are likely to experience the highest levels of well-being among the four types. Their ability to effectively leverage AI tools can lead to reduced stress, increased job satisfaction, and a strong sense of accomplishment. They may be best positioned to realize Keynes’ vision of increased leisure time without sacrificing productivity or career advancement. However, they may also face challenges in maintaining work–life balance due to their high achievement orientation.
Career preparation: Going beyond task performance
To fully understand the impact of AI on work hours, it is crucial to recognize that work involves not just task performance behaviors but also career preparation behaviors (Rynes & Gerhart, Reference Rynes and Gerhart1993). Generative AI is poised to make the task performance dimension more efficient, potentially freeing up time for career preparation activities. However, the reallocation of this saved time is likely to vary among different worker types.
Career preparation encompasses all activities aimed at skill development, learning, and positioning oneself for future career opportunities. This includes formal training, informal learning, networking, strategic planning, and staying abreast of industry trends. Although AI may reduce the time needed for task performance, conscientious workers, particularly AI augmenters, are more likely to invest the time saved into career preparation activities.
For these workers, overall work hours may not change significantly, even if they save considerable time on task performance. This aligns with research on conscientiousness and time use, which suggests that highly conscientious individuals tend to allocate more time to work-related activities. By investing in career preparation, these workers are likely to stay ahead of their peers and continue to bring the weekly work hours to the same 40-h equilibrium that exists today.
In contrast, surface adopters and laggards may be less inclined to reinvest time savings into career preparation activities. For these groups, the efficiency gains from AI could potentially lead to reduced work hours but at the cost of long-term skill development and career progression. This dynamic highlights a potential paradox in the AI-driven workplace: Although technology may enable reduced work hours, individual differences in personality and work attitudes may lead to divergent outcomes. Some workers may use AI to work less while maintaining productivity, whereas others may work the same hours but dramatically increase their output and skills.
In conclusion, although generative AI presents an opportunity to move closer to Keynes’ vision of reduced work hours, its impact is likely to be more varied and complex than a universal reduction in work time. The key to realizing the benefits of this technology while mitigating its potential downsides lies in fostering a workforce that combines a strong work ethic with high AI adaptability and in creating organizational systems that support continuous learning and skill development.
Typology of workers in the AI era.
