Ahead of World Day for Safety and Health at Work on 28 April 2025, the International Labour Organization published a global report, Revolutionizing health and safety: The role of AI and digitalization at work, on 23 April 2025. It covers automation and robotics, smart safety tools and monitoring systems, virtual and extended reality, algorithmic management of work, and changing work arrangements such as telework and digital platforms (International Labour Organization [ILO], 2025a).
In Indonesia, hazard reporting apps, gas sensors, smart cameras and virtual reality training are offered to companies more and more often. Below are six questions we often hear in the field, answered from the ILO report and other traceable sources.
What Is the Report?
Its main message is balanced. Robots can replace workers in dirty, dangerous and demeaning jobs, and automation can cut repetitive tasks. But those benefits only come if the technology is introduced without creating new risks (ILO, 2025b). At workplace level, the ILO stresses regular risk assessment, controls that follow the hierarchy of controls, and the principle that AI analytics and real-time monitoring must complement, not replace, human judgement (ILO, 2025a).
How Widespread Is It?
Less widespread than the marketing suggests. A review for the European Agency for Safety and Health at Work (EU-OSHA) cites the ESENER-3 survey: only 4.8 percent of establishments surveyed used wearable devices and 3.7 percent used collaborative robots, far below PCs at fixed workplaces (86.6 percent) and mobile devices (76.7 percent). By contrast, 11.8 percent used systems to set the content and pace of work, and 8.2 percent used systems to monitor worker performance (Andriescu et al., 2022). These are European data and cannot be applied directly to Indonesia. In our view the figures also show that systems to manage and monitor performance are more common than safety devices, so companies need to keep the two apart from the start so that safety goals do not get mixed with performance appraisal.
What Benefits Are Shown?
The ILO report includes several examples. In a 12-month trial in 2021 on a high-speed railway project in the United Kingdom, smart ear defenders recorded at-ear noise exposure and how the protection was used. Changes made from those data contributed to a 50 percent reduction in overall noise exposure. Whole-body powered exoskeletons have been reported to cut back muscle activity by up to 53 percent and leg strain by up to 63 percent (ILO, 2025a). In our view these are strong illustrations, but they come from specific trials, so results elsewhere may differ.
What New Risks Arise?
The ILO notes several. System failures or data transmission errors can delay alerts and raise accident risk when workers rely too heavily on automated warnings. Devices that are uncomfortable or fit poorly tend not to be worn. Constant monitoring of physiological data can cause stress, and it raises questions about consent, data security and misuse of workers' health data. In virtual reality, cybersickness is reported to affect 20 to 80 percent of users (ILO, 2025a).
Can AI Read Emotions?
In the European Union, the AI Act prohibits the use of AI systems to infer a person's emotions in the workplace, except for medical or safety reasons, under Article 5(1)(f). The prohibitions in Chapter II have applied since 2 February 2025. AI systems used to monitor and evaluate the performance and behaviour of workers are classed as high-risk under Annex III, point 4 (European Parliament & Council of the European Union, 2024). The rule does not apply directly in Indonesia, but in our view it is a useful reference for companies serving European markets or reporting to a European parent.
How to Start Safely?
EU-OSHA distinguishes proactive systems, which prevent harm before an accident, from reactive systems, which support emergency response and accident recording. Both need to be seen within one cycle of safety improvement (Andriescu et al., 2022). From there, sensible steps are:
- Start from the risk, not the tool. Define the hazard to be controlled, then judge whether a digital tool is the right control.
- Involve workers and their representatives in selection, setting the purpose, introduction and review of the tool, as the ILO recommends (ILO, 2025a).
- Decide at the outset what data are collected, who can see them, and that safety data will not be used to rate individual performance.
- Prepare a procedure for system failure so workers do not depend entirely on automated warnings.
- Train users and supervisors, then check whether safety indicators actually improve.
Nusa Safety Support
Nusa Safety can help design risk-led digitalization through its HSE digitalization and custom HSSE software services, and prepare user training material through custom training module development. Examples of applications are on the HSSE Apps page. Nusa Safety issues internal training certificates only.
This article is general information and not legal or professional advice for any specific case.
References
- Andriescu, M., Battaglini, M., Spyridopoulos, K., Kilic, L., Olausson, N., Broughton, A., & Toro, D. (2022). Smart digital monitoring systems for occupational safety and health: Uses and challenges. European Agency for Safety and Health at Work. https://doi.org/10.2802/315859
- European Parliament & Council of the European Union. (2024). Regulation (EU) 2024/1689 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act). Official Journal of the European Union, L series. https://eur-lex.europa.eu/eli/reg/2024/1689/oj
- International Labour Organization. (2025a). Revolutionizing health and safety: The role of AI and digitalization at work. https://www.ilo.org/publications/revolutionizing-health-and-safety-role-ai-and-digitalization-work
- International Labour Organization. (2025b, April 23). AI and digitalization are transforming safety and health at work [Press release]. https://www.ilo.org/resource/news/ai-and-digitalization-are-transforming-safety-and-health-work
