Displacement, not replacement: how AI will affect workers, according to Goldman Sachs

Widespread AI adoption could displace 6-7 per cent of the US workforce, new research by Goldman Sachs has found. However, researchers anticipated that AI-related job losses would be fleeting and “frictional,” as workers re-trained and adapted to technological shifts.

“A recent pickup in AI adoption and reports of AI-related layoffs have raised concerns that AI will lead to widespread labor displacement,” the report’s authors Joseph Briggs and Sarah Dong noted.

“While these trends could broaden as adoption increases, we remain skeptical that AI will lead to large employment reductions over the next decade.”

The research said that AI could hypothetically lead to an increase in unemployment through two key channels - long-term structural unemployment, and temporary frictional unemployment.

Firstly, AI capabilities could advance so much that human input would become redundant, leading to structural unemployment.

Goldman Sachs saw this outcome as unlikely, as technological change historically boosted demand for workers in new occupations. For reference, 60 per cent of US workers today perform occupations that did not exist in 1940.

Secondly, AI could boost frictional unemployment, a common phenomenon during technological shifts where workers’ skills become mismatched to the demands of the labour market, forcing them to pivot or retrain in more relevant skills.

Goldman Sachs revealed that there was early evidence that generative AI was disproportionately affecting young tech workers. Unemployment among 20-30 year olds in tech-exposed occupations rose by almost 3 percentage points since the start of 2025.

Anecdotal reports also suggested generative AI was causing hiring headwinds for recent technology graduates.

“While some of the decline can be attributed to payback from over-hiring during the pandemic, tech’s employment share has now fallen below its (remarkably linear) pre-pandemic trend, a pattern probably related to AI automation,” the authors noted.

The survey noted that less than 2.5 per cent of US employment was at risk of automation. Financial specialists, customer service representatives, software developers, accountants and auditors were among the professions at highest risk of displacement.

Those least at risk of displacement were air traffic controllers, chief executives, radiologists, pharmacists, residential advisors and photographers.

While Goldman Sachs predicted that AI would have limited impacts on long-term employment levels, the researchers acknowledged that AI adoption remained in its early stages. They noted that its impact on jobs would be largely dependent on how employers put the technology to use.

“Until the AI adoption cycle has fully played out, the potential labor market disruption—including which jobs are likely to be displaced by generative AI—will remain an open question,” Briggs and Dong wrote.