Korea could see 3.41 mn jobs replaced by AI technology: BOK analysis

[Image source: Gettyimagesbank]

South Korea could see about 3.41 million domestic jobs replaced by artificial intelligence (AI) as the technology continues to spread across industry sectors, according to an analysis by the Bank of Korea (BOK) on Thursday. The report specified that doctors, accountants, and lawyers, classified as typical high-income occupations, are expected to be hardest hit.

According to a BOK report published on Thursday, which analyzed the potential impact of AI on jobs and the labor market by using AI patent information, 12 percent of all jobs, or 3.41 million jobs, are at high risk of being replaced by AI technology. It added if the technology is expanded further moving forward, the number of at-risk jobs could increase to 14 percent, or 3.98 million.

The analysis adapted the concept of an ‘AI exposure index’ from a paper by Stanford University professor Michael Webb, converting it to the Korean Standard Classification of Occupations to identify jobs highly exposed to AI in the domestic context. The index was created by examining the number of AI patents in each occupation to measure the degree to which a particular job can be replaced by AI technology.

Jobs with the highest AI exposure included chemical engineers, power plant operators, railroad and subway operators, and water and wastewater treatment operators. These jobs involve tasks suitable for using large-scale data to streamline their work. By occupational classification, high-income professions such as doctors, accountants, asset managers, and lawyers were also identified as occupations susceptible to being replaced by AI.

Jobs that are difficult to replace due to their low AI exposure index, on the other hand, include those that require face-to-face contact and relationship building, such as food-related workers, university professors and lecturers, and religious workers. Additionally, occupations like journalists, singers, and vocalists were deemed less replaceable by AI.

When examined by wage and education levels, highly educated and high-income workers were more exposed to AI. This contrasts with how previous innovative technologies, such as industrial robots and software, had a significant impact on less educated, middle-income workers. For example, the replacement rate for industrial robots was high in the order of high school graduates, college graduates, and graduate school graduates, but was reversed for AI. This suggests that the higher the AI exposure index, the greater the likelihood of lower employment and lower wage growth.

Based on the results of the decline in the employment rate and wage growth of jobs with a high exposure to industrial robots and software between 2000 and 2021, the central bank analyzed that the employment rate for jobs with a high AI exposure index fell by 7 percentage points and wage growth by 2 percentage points.

“Productivity gains from AI can lead to an overall increase in labor demand and higher wages,” according to BOK. “But some workers could face difficulties in the job transition process brought by the introduction of AI, as the productivity effect affects the entire economy, but the substitution effect is concentrated in certain groups.”

In this regard, workers are likely to be rewarded more for ‘soft skills’ including social skills, teamwork, and communication abilities, which will increasingly be required in the changing job landscape.