AI Job Shock Hits Young Workers First as Goldman Finds Entry-Level Jobs Face 3 Times Greater Employment Drag

AI is hitting entry-level workers far harder than the rest of the labour market, with Goldman Sachs finding that jobs at the start of a career suffer more than three times the employment drag from artificial intelligence as the wider workforce.

The findings come alongside fresh figures from South Korea and Britain that point to sharp falls in youth employment in sectors most exposed to the technology.

Goldman Sachs analysed more than 800 occupations across developed economies and found that the employment growth drag associated with AI exposure was more than three times larger for entry-level workers than for the workforce overall.

The research comes as the Bank of Korea reports that 94% of youth employment losses in South Korea since 2022 occurred in sectors with high AI exposure. In Britain, youth unemployment has risen to 16.4%, its highest level in about a decade.

Goldman Finds Larger Impact on Entry-Level Workers

Goldman Sachs found that a 10% increase in occupational exposure to AI was associated with a reduction in annual headcount growth among entry-level workers that was more than three times larger than the effect across the broader workforce. The analysis covered more than 800 occupations across developed economies.

The research also found significant employment declines in some industries with high AI exposure. US call-centre employment was 39% below its historical trend, while employment in the sector was 33% below trend in Canada and 27% below trend in Germany.

The effects were concentrated in particular industries rather than being spread evenly across the labour market, according to the research. Goldman Sachs also noted that the relationship between AI exposure and employment does not establish that AI was solely responsible for changes in headcount.

South Korea Reports Sharp Youth Employment Decline

The Bank of Korea reported that employment among people aged 15 to 29 fell by 285,000 between June 2022 and June 2026.

Of those losses, 268,000, or 94%, occurred in industries with high exposure to AI, according to the central bank's analysis.

The largest employment declines in AI-exposed sectors were recorded in information services, publishing, computer programming and professional services.

Information services employment fell 31.4%, publishing declined 27.4%, computer programming dropped 16.6% and professional services fell 11.6%.

The Bank of Korea also reported a rise in employment among workers in their 50s in the same high-exposure industries. Employment among that age group increased by 230,000 during the period, with 173,000 of those gains occurring in highly AI-exposed sectors.

The central bank said the findings pointed to a shift towards a more seniority-biased labour market. It also found that the impact varied according to how companies used AI.

Employment effects were more pronounced in industries where AI was used for automation, while the impact was more limited where the technology was used to assist workers.

UK Youth Unemployment Reaches 16.4%

Britain has also recorded a rise in youth unemployment. There were 743,000 unemployed people aged between 16 and 24 in the three months to May 2026, according to data from the Office for National Statistics cited by the House of Commons Library. That was 112,000 more than a year earlier.

The youth unemployment rate rose to 16.4%, compared with 14.2% a year earlier. The rate was the highest recorded since the period covering July to September 2015. The number of young people aged 16 to 24 who were not in employment, education or training also rose.

There were 957,000 young people in that category in the first quarter of 2026, according to the House of Commons Library. Britain's overall unemployment rate was 4.9% between March and May, while the number of vacancies fell to 712,000 between April and June, according to the latest labour-market figures.

AI Exposure Remains Concentrated

The Goldman Sachs research found that employment effects were concentrated in occupations with higher levels of AI exposure rather than across the entire labour market. The analysis included sectors such as software publishing, management consulting, advertising and call centres.

The Bank of Korea similarly found differences between industries where AI was primarily used to replace tasks and those where it was used to augment human work.

The research from both institutions therefore focuses on employment patterns in AI-exposed occupations and sectors rather than establishing a direct causal relationship between AI adoption and overall youth unemployment.

Goldman Sachs' findings add to a growing body of research examining how the technology is affecting workers at different stages of their careers, while the South Korean data provides a separate measure of youth employment changes in industries with high AI exposure.