Around 300 million jobs worldwide are exposed to automation from artificial intelligence, a shift that could reshape labour markets over the next decade even as the technology lifts productivity and creates new roles, according to analysis from Goldman Sachs Research.
The impact of AI is already visible in pockets of the US economy, particularly in technology, said Joseph Briggs, who co-leads the bank’s global economics team. Employment in the tech sector has fallen below its long-term trend, a move Briggs partly attributes to early adoption of AI tools.
In Goldman Sachs Research’s base-case scenario, it could take about 10 years for firms to adopt AI on a wide scale, with roughly 6-7 per cent of workers displaced during that transition. If the shift occurs gradually, the US unemployment rate could rise by about 0.6 percentage points. A more front-loaded transition, Briggs said, would have a far larger economic impact.
So far, job displacement linked to AI has been limited to relatively small parts of the labour market, including tech roles and some knowledge and creative professions such as management consulting, call centres and graphic design. Broad labour data have yet to show significant AI-driven changes in the overall employment mix across the US economy.
Looking ahead, the effects are expected to widen. In the United States alone, AI could automate tasks accounting for about 25 per cent of total work hours, Goldman Sachs Research estimates. Workers displaced from knowledge-intensive roles may struggle to move into the kinds of jobs most in demand, said Evan Tylenda, an analyst at GS SUSTAIN.
Those needs span both low-wage service work, such as cleaning and home healthcare, and higher-skill technical roles including construction, engineering and electrical work. In the US, around 5,00,000 net new jobs will be required by 2030 to meet rising power demand linked to AI-driven data centre expansion, Tylenda said.
Hiring trends tied to that build-out are already emerging. Employment for HVAC contractors, electrical contractors and other construction workers involved in data centre projects has increased by about 2,16,000 since 2022, Briggs said, adding that investment in such infrastructure is expected to keep growing.
Beyond infrastructure, Briggs said the AI revolution is likely to create new kinds of employment, including rising demand for workers with AI skills, the emergence of specialised occupations in areas such as healthcare, and indirect job growth driven by higher incomes and consumer demand. Similar forces have shaped the labour market in the past, he noted, pointing to roles such as pet care, tutoring and athletic coaching, which together employ around 1 million workers today and largely emerged over the past three decades.
