Author: Hilman Nurjaman

Editor: Ayom Mratita Purbandani

As artificial intelligence (AI) becomes more present in everyday life, including in the workplace, many companies have started using algorithmic management systems to organize activities such as scheduling, setting targets, and monitoring workers. An OECD survey conducted in 2023 shows that workers who interact with algorithmic management often report higher work intensity, stronger pressure to meet targets, and growing concerns about privacy.[1][2] At the same time, the use of algorithms can narrow workers’ autonomy as individuals.[3] In this context, this commentary explores how the implementation of algorithmic management may intensify work while reducing workers’ autonomy, with potential negative consequences for job quality and worker well-being.

The diagram above shows changes in the level of job autonomy perceived by workers who use AI, particularly in terms of their control over the order in which they complete their tasks. The data indicate that some workers who are directly managed by AI report a decline in their level of autonomy, suggesting that algorithm-based management can reduce workers’ control over how they organize and carry out their work.

Why is algorithmic management becoming more popular?

One reason algorithmic management is becoming increasingly popular is its promise of efficiency and consistency. A report from the OECD shows how widely algorithmic systems have already been adopted in the workplace, around 90% of companies in the United States and 79% in Europe have used at least one type of algorithmic tool.[4] These tools can automate many managerial tasks that were previously handled by humans, such as scheduling shifts, assigning tasks, and evaluating worker performance,[5] companies often see them as a way to make decisions faster and, supposedly, more objective. A well-known example is Amazon, which has implemented algorithmic management quite extensively. One of its systems, known as Time Off Task (TOT), automatically tracks workers’ idle time. If the system detects 30 minutes without activity, it can trigger a warning, one hour may lead to disciplinary action, and two hours can even result in termination through an automated process.[6] Similar systems have also been adopted by many other large companies, replacing part of the role traditionally played by human managers in strategic and administrative matters.

Unrealistic targets, rising stress, and declining worker autonomy

The use of algorithmic systems in management often pushes workers to work harder in order to meet performance targets set by the system. A report from the OECD notes that constant monitoring and automated, data-driven performance evaluations can create a work environment that encourages stress.[7] When every activity is tracked and assessed by a system, workers can feel as if they are constantly being watched and pressured to meet ever-increasing targets. In delivery work, for example, algorithmic AI systems often push drivers to “beat their time,” encouraging them to complete deliveries faster than before. While this may improve efficiency, it can also reduce worker safety as drivers rush to meet the targets set by the system. A similar dynamic appears in warehouse operations, where algorithmic management tools continuously assign picking targets to workers. Combined with the possibility of disciplinary action or termination, these targets can lead to excessive work intensification, rising stress levels, and a greater risk of both physical and mental exhaustion.[8]

A similar pattern can also be seen among gig workers across Southeast Asia. Field studies show that drivers for ride-hailing platforms and online couriers are often pushed to work long hours with very limited breaks in order to meet targets determined by algorithmic systems.[9] Many drivers report accepting trips they would normally avoid, hoping that the algorithm will reward them with bonuses once they reach certain performance thresholds. Long working hours and insufficient rest have therefore become common, contributing to higher levels of fatigue, anxiety, and negative impacts on both mental health and physical safety.

Continuous monitoring and automated task allocation leave little room for dialogue, negotiation, or human judgment in adjusting how a task should be carried out. In more extreme cases of automation, warehouse workers may no longer be able to make even small decisions about how to perform their jobs, such as choosing a more efficient way to move items, because every movement is directed by an algorithm.[10] This high level of surveillance can make workers feel less like individuals and more like extensions of the system itself, leading to feelings of alienation and a loss of emotional engagement with their work. The lack of transparency in these systems can further deepen frustration, as workers often do not know the logic behind the decisions that affect them and have little opportunity to challenge targets or sanctions imposed by the system. For example, at Amazon, algorithm-based performance monitoring systems are known to automatically issue warnings and can even trigger worker termination when productivity levels are considered too low.[11] Yet workers typically do not have access to the calculations behind these evaluations or the ability to correct possible errors made by the algorithm. Situations like this can create a strong sense of job insecurity, where a system error or miscalculation may lead to disciplinary action, or even job loss, without a transparent review process.

Consequences for job satisfaction and worker inclusivity

The effects of work intensification and tight control can also be seen in workers’ levels of job satisfaction. Workers who are heavily managed by algorithms often report higher work intensity, reduced job autonomy, and growing concerns about privacy.[12] Therefore, many of them feel more pessimistic about the impact of AI on their jobs. Constant pressure and the loss of autonomy can undermine mental well-being, as workers may feel continuously under scrutiny and less trusted in how they perform their work. From the perspective of inclusivity, AI can produce mixed outcomes. On the one hand, certain AI technologies can expand access to work for people with disabilities. Tools such as speech recognition or automatic text generation, for instance, can make communication easier and help individuals participate more fully in the workplace.[13] On the other hand, algorithmic systems that operate without sufficient oversight may reinforce or even widen existing inequalities. Biased algorithms can disadvantage vulnerable groups, such as older workers or those with lower levels of digital skills, limiting their access to fair opportunities and working conditions.

At a broader structural level, algorithmic management also reflects a shift in power. Algorithms (designed and controlled by companies or digital platforms) are increasingly taking the central role in directing work systems. In many ways, this resembles a new form of digital Taylorism, where algorithms regulate the details of tasks so closely that workers risk becoming little more than “appendages of the machine,” with limited autonomy over their work.[14] Algorithmic surveillance reduces the need for direct personal supervision, while the role of traditional managers shifts toward verifying and interpreting system-generated data. A clear example can be seen in ride-hailing platforms. Drivers are often labeled as “independent contractors,” yet the algorithm determines task allocation, deciding when, where, and which rides they receive. Because of this “independent contractors” status, many protections normally associated with formal employment, such as safety guarantees or minimum wages, may be absent. This creates a paradox of accountability, workers are expected to bear the consequences of their performance, yet they have little influence over the decision-making processes that shape their work in the first place.

Conclusion

The broader picture suggests that the rise of algorithmic management often leads to higher work intensity and reduced worker autonomy, with negative consequences for job quality and workers’ well-being. Examples from different industries and online platforms illustrate how strict performance targets and constant digital monitoring can gradually erode workers’ satisfaction and health. For this reason, the productivity and efficiency promised by AI should not come at the expense of its social consequences. Companies and policymakers need to take a closer look at how these systems operate and ensure clearer transparency, accessible complaint mechanisms, and stronger protections for workers’ rights and safety. Without such safeguards, algorithmic management risks prioritizing efficiency over fairness and well-being in the workplace.

1. OECD. (2023). OECD Employment Outlook Artificial Intelligence and The Labour Market. ↑

2. OECD. (2025). Algorithmic Management in The Workplace: New Evidence from an OECD Employer Survey. OECD Artificial Intelligence Papers. ↑

3. OECD. (2023). OECD Employment Outlook Artificial Intelligence and The Labour Market. ↑

4. OECD. (2025). Algorithmic Management in The Workplace: New Evidence from an OECD Employer Survey. OECD Artificial Intelligence Papers. ↑

5. OECD. (2025). Algorithmic management in the workplace. ↑

6. Kalluri, R. (2026). The Algorithmic Middle Manager: Are We Building Managers or Checkers for Corporate America?. California Review Management. ↑

7. OECD. (2023). OECD Employment Outlook Artificial Intelligence and The Labour Market. ↑

8. Moore, P. (2018), The Threat of Physical and Psychosocial Violence and Harassment in

   Digitalized Work, ILO, Geneva. ↑

9. Tan, J. & Gong, R. (2024). The Plight of Platform Workers Under Algorithmic Management in Southeast Asia. Carnegie Russia Eurasia Center. ↑

10. Brione, P. (2020), My boss the algorithm: an ethical look at algorithms in the workplace, Acas. ↑

11. Soper, S. (2021). Fired by Bot: Amazon Turns to Machine Managers and Workers Are Losing Out, Bloomberg. ↑

12. OECD. (2023). OECD Employment Outlook Artificial Intelligence and The Labour Market. ↑

13. Smith, P. & L. Smith (2021), Artificial intelligence and disability: too much promise, yet too

    little substance?. AI and Ethics 1(1), pp. 81-86. ↑

14. Staab, P. & Nachtwey, O. (2016). Market and Labour Control in Digital Capitalism. Communication Capitalism & Critique Open Access Journal for a Global Sustainable Information Society, 14(2) pp. 457-474. ↑