The effect of AI adoption is positive and highly statistically signifi-cant for STEM workers. By contrast, it is negative and only marginally not significantfor non-STEM employees

positive but imprecisely estimated coefficient for high-skill workers shown in column (1).These results suggest a specific aspect of upskilling in a period of rising deployment ofAI.

The largest negative effect is found for middle-age work-ers, which account for almost 88% of the effect on overall employment

Figure 1. Employment Share of AI-Related Occupations in the US

decennial Census for the year 2000 and theAmerican Community Survey (ACS) for the years 2010 and 2020.

For in-stance, Hui et al. (2023) document a negative impact of generative AI on the employ-ment of free-lancers in an online labour market; Grennan and Michaely (2020) findnegative effects for financial analysts; Armour et al. (2022) find evidence of lawyers beingdisplaced by AI; and Abis and Veldkamp (2024) estimate that changes in data intensityled to a 5% decline in the labour share within the investment management industry.

On the one hand, AI can sometimes augmentrather than substitute workers.

We show that thenegative employment effects are largest for low-skill and production workers, while theyturn strongly positive for workers in the top decile of the wage distribution and for occu-pations requiring a STEM degree.

In all cases, we estimate robust negative effects of AI exposure on employmentacross CZs and time.

. Whether AI willcomplement or substitute workers is therefore an empirical question, for which there isstill little systematic evidence

hniques and the growing availability of vast amounts of digital data, the last two dec-ades have witnessed a tremendous increase in the use of AI applications, which includeweb search engines, targeted advertising, recommendation systems, self-driving cars,generative or creative tools and chatbots.

Collaboration with artificial intelligence technologies:

Artificial intelligence governance (regulation and transparency):

Governments should encourage policies that promote job retraining andupskilling programs.

To sum up, artificial intelligence innovations can affect employment and jobsecurity by directly displacing workers from tasks, they were previously performing(displacement effect). Additionally, by increasing the demand for labor in industries or jobsthat arise or develop due to technological progress (productivity effect). (Petropoulos, 2018,p. 119)

Georgios Petropoulos argues that middle-level jobs thatrequire routine manual and cognitive skilled are the ones that are most at risk, but jobs thatemphasize interpersonal skills are much harder to be replaced by AI. ( Neufeind, O’Reilly,& Ranft, 2018, p. 7).

Ifthe organization does not take the corresponding adjustment measures in time, theemployee's anxiety will have a negative impact on the organization. ( Liu & Zhan, 2020, p.2).

Many researchers are concerned that AI will result in more inequality due to its threatof disrupting labor markets. It is argued that the continuous growth in automation, robotsand computers will take the jobs of workers in many industries with the most worryingfactor being the increased danger of the disappearance of low-skill/low-pay jobs whichwill cause a lot of challenges for the poor which will lead to a rise in social tension.

In manufacturing, for example, in order to fulfill customized orders and handlefluctuations in demand, employees are increasingly collaborating with robots to perform newtasks without having to manually overhaul any processes. AI technologies are also performingcertain tasks autonomously

Figure 01 : Objectives of AI Applications

30% of the organizations that are listed as 30 mostsuccessful organizations in the world are using some kind of AI.

Surveys indicate that less than halfof large organizations have meaningful AI projects in process, although the percentage isgradually increasing . ( Benbya & et al, 2020, p. 9).

“the science and engineering of making intelligent machines, especially intelligentcomputer programs. It is related to the similar task of using computers to understand humanintelligence”

concerns thatthe increasing sophistication of artificial intelligence may negatively affect the labor forceand lead to mass unemployment by jeopardizing skilled and semi-skilled workers and reducethe size of the middle class, have made AI one of the parts of the debate on the link betweenautomation and jobs, which gives rise to concern about a massive and rapid disappearanceof the latter.

The Fourth Industrial Revolution (4IR) or “Industry 4.0”