{
  "id": 5397597,
  "title": "How AI could make us work even harder",
  "url": "https://urgent.news/2026/09/03/how-ai-could-make-us-work-even-harder",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-03T18:20:04.000Z",
  "source": {
    "name": "Phys.org",
    "slug": "phys-org",
    "url": "https://phys.org/news/2026-09-ai-harder.html"
  },
  "original_language": "en",
  "account": "Recent technological advancements have been taking a significant amount of workload off humans, with machines handling tasks like washing clothes, plowing fields, and sorting administrative work. Economic growth and wealth redistribution have resulted in most workers in developed countries not needing to dedicate most of their waking hours to paid employment. The 12-hour workday prevalent in the 19th century has largely decreased. However, the introduction of Artificial Intelligence (AI) may not lead to more free time, and could even result in employees working even harder.\n\nOne reason for this potential increase in work hours is the impact of AI on workers' bargaining power in the workplace. If employees are concerned about job loss due to automation, research suggests they may be more inclined to increase their efforts as a way to communicate their value to employers. Studies indicate that this response is expected, as employers need to decide on hiring, promoting, or dismissing workers, aiming to reward productivity. However, measuring productivity can be challenging, leading employers to rely on metrics such as the amount of time spent working. Checking if a workstation remains occupied after all others have left for the day is easier than assessing the accuracy of complex projects. Increasing the amount of time spent at work seems like a logical way for workers to distinguish themselves from others who may appear less motivated.\n\nThe introduction of AI to the workplace can exacerbate this dynamic. Existing AI tools have already automated certain tasks, like data entry, creating an expectation that entire roles might soon become redundant. This context intensifies the competition among employees, who may suspect large-scale restructuring is imminent. Evidence shows that workers in industries exposed to AI, such as financial services and IT, are staying later at work than those in less exposed industries. Another issue is that employers may overestimate the extent to which AI enhances workers' productivity. The availability of AI tools for increased efficiency leads employers to expect workers to accomplish more in the same amount of time. However, if these expectations are unrealistic, they could lead to additional work intensification or even longer work hours.\n\nLastly, the adoption of AI can cause significant disruptions. Established routines are disrupted, existing roles change, and new roles emerge. Workers must be trained on how to use AI tools and learn to work alongside them. However, this is particularly challenging as AI tools are continuously improved and new tools developed rapidly. All these factors place additional demands on workers' time. While it may be disheartening to consider that AI might not alleviate our labor burdens, this situation is not necessarily inevitable or unavoidable. By controlling AI's design and implementation, employers and large AI firms could create a workforce that is empowered, enhancing their skills and making their work more valuable. This, in turn, would strengthen their bargaining power with employers, allowing for better working conditions, including more flexible work hours that align with individuals' lives and preferences. If developers prioritize augmentation and empowerment, AI could potentially be the ultimate tool for tackling widespread resentment and disenchantment with modern working lives.",
  "summary": "Technological progress has taken more than its fair share of the workload from humans over the years. Washing machines clean our clothes, tractors plow the fields and computers sort out administrative work.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}