Artificial intelligence is beginning to leave visible marks on the labor market. For now, the effects are not seen everywhere, but in some fields, the number of employees has decreased significantly below the trend of recent years. And those at the beginning of their careers seem to be the most exposed.
The conclusions come from an analysis by Goldman Sachs on labor markets in major developed economies, cited by CNBC.
The investment bank found that sectors in which a larger part of the activity can be automated with the help of AI have generally recorded a slower growth in the number of job opportunities available starting from the second half of 2022. The phenomenon is most visible in Germany, Australia, and the United States.
In the information and communications sector, one of the most exposed to artificial intelligence, the growth in the number of employees has slowed down in almost all major developed economies since 2022.
However, there is an important difference: outside the United States, the number of employees in this sector remains close to or even above the long-term trend.
Call centers, among the most affected
The picture becomes clearer when analyzing separately the sectors where AI can already take over a significant part of the tasks previously done by people.
Goldman Sachs shows that the number of employees in call centers, the software industry, management consulting, and advertising has dropped significantly below the historical trend in developed economies.
The biggest difference is seen in call centers. In the United States, the number of employees in this sector is 39% below the level indicated by the historical trend. In Canada, the difference is 33%, and in Germany, it is 27%.
According to Goldman Sachs, these figures suggest that the effects of AI on jobs are already becoming visible where there are tools that can automate some of the employees' tasks.
However, this does not mean that artificial intelligence is already causing massive job losses across the entire economy. The analysis rather shows that the effects are currently concentrated in a relatively small number of sectors.
Those at the beginning of their careers are more vulnerable
One of the most important conclusions of the analysis concerns young people trying to enter the labor market.
Goldman Sachs analyzed the evolution of the number of employees in over 800 occupations and found that the pressure associated with artificial intelligence is stronger for entry-level jobs.
The explanation also lies in the type of work done in the early years of a career. Many of the repetitive or standardized tasks that were previously assigned to employees with little experience are precisely the ones that current AI tools can more easily take over.
The numbers show how significant the difference is. Across the entire labor market, a 10% exposure of an occupation to AI is associated with only a 0.1 percentage point reduction in the annual growth rate of the number of employees in France, Canada, and the USA.
For employees at the beginning of their careers, the effect is greater: exceeding 0.6 percentage points in Australia and 0.2 percentage points in the United States.
Goldman also identified a negative effect in occupations considered particularly vulnerable to replacement by AI, but this is more limited.
The bank's conclusion is that the pressure of AI on employment can already be observed in data from several countries, but it is not, at least for now, a widespread phenomenon.
AI is already used by up to 20% of developed economies
The changes in the labor market come as artificial intelligence is being adopted more rapidly.
To see how widespread the technology is, Goldman Sachs combined the results of 11 surveys conducted in several countries. The conclusion: in the main developed economies, the adoption rate of AI is currently around 15%-20%.
Leading the way are France, the United States, the Netherlands, and the United Kingdom. At the other end of the ranking, among developed economies where AI is less used, are Italy, Japan, and New Zealand.
In major emerging economies, the estimated adoption rate is somewhat lower, between 10% and 15%.
Goldman Sachs' data thus suggest that AI is not currently causing a general shock in the labor market, but the change has already begun. And the first effects are seen precisely in sectors where technology can quickly take over tasks previously done by people and among those trying to find their first job.
G.P.