AI shows little impact on worker well-being despite self-reported job satisfaction concerns

While artificial intelligence reshape the workplaces around the world, a new study provides early evidence suggesting that AI exposure has not, so far, caused damage to mental health or workers’ work satisfaction. In fact, data reveal that AI can even be linked to modest improvements in workers’ physical health, especially among employees with less than a university diploma.

But the prudent authors: it is too early to draw definitive conclusions.

The article, “Artificial Intelligence and the Wellbeing of Workers”, published on June 23 in Scientific relationshipsUse two decades of longitudinal data from the German socio-economic panel. Using these rich data, researchers – USA Giuntella from the University of Pittsburgh and the National Bureau of Economic Research (Nber), Luca Stella of the University of Milan and Berlin School of Economics, and Johannes, the King of the German Ministry of Finance – How Workers in Expose A -Exposed occupations was heard in relation to workers in less roles Exposed.

“Public anxiety with regard to AI is real, but the worst scenarios are not inevitable,” said Professor Stella, who is also affiliated with independent European organizations The Center for Economic Studies (CESIFO) and the Institute for Labor Economics (IZA).

“So far, we find little evidence that the adoption of AI has undermined the well-being of workers on average. If anything, physical health seems to have improved slightly, probably due to the decline in the physical intensity of work and the overall risk of employment in some of the professions exposed to AI.”

However, the study also highlights the reasons for prudence.

The analysis is mainly based on a measure based on the tasks of exposure to AI – considered to be more objective – but alternative estimates based on self -disclary exposure reveal small negative effects on work and life satisfaction. In addition, the sample excludes young workers and only covers the first phases of the dissemination of AI in Germany.

“We can simply be too early in the AI ​​adoption curve to observe its complete effects,” said Stella. “The impact of the AI ​​could evolve considerably as technologies are progressing, enter more sectors and modify work to a deeper level.”

The main results of the study include:

• No significant average effect of exposure to AI on work satisfaction, life satisfaction or mental health.
• Small improvements in self -evaluated physical health and health satisfaction, especially in low education workers.
• Proof of reduction in the physical intensity of work, which suggests that AI can mitigate physically demanding tasks.
• A modest drop in weekly working hours, without significant variation in income or employment rates.
• Self-declaced exposure to AI suggests weak but negative effects on subjective well-being, strengthening the need for more granular future research.

Due to the supply of data, the study focuses on Germany – a country with strong work protections and a progressive pace of AI adoption. The co-authors noted that the results can differ in more flexible labor markets or among the younger cohorts entering the increasing saturated workplaces.

“This research is an early snapshot, not the last word,” said Giuntella de Pitt, who previously carried out important research on the effect of robotics on households and work, and on the types of workers.

“As the adoption of AI accelerates, continuous monitoring of its broader impacts on work and health is essential. Technology does not determine the results – institutions and policies will decide whether AI improves or erodes working conditions.”