Goldman Sachs warns AI could automate 25% of work hours as white-collar jobs face disruption

Reviewed byNidhi Govil

2 Sources

Share

Goldman Sachs projects AI will automate nearly a quarter of total work hours, with white-collar and knowledge-intensive roles most exposed. The investment bank predicts a temporary rise in unemployment by 0.6 percentage points but emphasizes that reskilling and policy response will determine whether automation leads to lasting job displacement or productivity gains.

Goldman Sachs Projects Major Work Automation Shift

A new report from Goldman Sachs has sent ripples through the employment landscape, projecting that AI could eventually handle nearly a quarter of total work hours across the global workforce

2

. The investment bank's analysis, which draws on labor data from the US Department of Labour, warns of significant workforce changes ahead while stopping short of predicting a widespread employment crisis. The 25% work hours automation estimate represents a substantial AI shockwave for industries that have long relied on human cognitive labor, though the impact will vary considerably by sector and occupation.

Source: Analytics Insight

Source: Analytics Insight

White-Collar Roles Bear the Brunt of Automation Potential

Office-based sectors and knowledge-intensive sectors sit at the center of this transformation. White-collar roles automation appears most imminent in positions requiring repetitive cognitive work, with clerical duties, data processing, entry-level coding, accounting, and legal research among the most exposed

2

. Administrative work, customer support, finance operations, and parts of engineering and design show high automation potential as generative AI impact expands

1

. These systems are already capable of composing papers, programming, and contract management on a large scale. Conversely, job roles demanding physical presence, manual skills, or human decision-making—such as construction, medical treatments, and vocational trades—remain fairly resistant to automation.

Source: Digit

Source: Digit

Job Displacement Concerns and Productivity Gains

Goldman Sachs does not predict mass job losses overnight, emphasizing that the estimate refers to work hours rather than entire roles

1

. Most professions consist of multiple tasks, and AI is anticipated to replace humans in performing tedious, repetitive tasks while workers move to more valuable activities. The research project suggests AI has the potential to increase labor productivity boost by approximately 15% over time

2

. Based on historical trends linking productivity gains to workforce changes, the report predicts approximately 6 to 7 percent of jobs may be lost during the adoption phase.

Temporary Rise in Unemployment Expected

Short-term disruption appears likely, with the job displacement potentially pushing unemployment up by about 0.6 percentage points at its peak—equivalent to roughly one million additional unemployed workers before new opportunities begin to materialize

2

. This temporary rise in unemployment underscores the transition period that workers and companies will navigate as AI take over jobs in certain sectors transforms from possibility to reality. Many sectors will be transformed before they are eradicated, creating a period of adjustment rather than immediate collapse.

Reskilling and Policy Response Prove Critical

The long-term impact of AI on employment will depend heavily on how quickly employees, companies, and policymakers respond to these shifts

2

. Investment in reskilling, education, and development of new AI-enabled roles will prove critical in determining whether work automation leads to lasting negative impacts or opens pathways to higher-value work. The Goldman Sachs analysis makes clear that the coming years will test society's ability to adapt to technological change while supporting workers through the transition.

Today's Top Stories

TheOutpost.ai

Your Daily Dose of Curated AI News

Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Spend less time searching for the latest in AI and get straight to action.

© 2026 Triveous Technologies Private Limited
Instagram logo
LinkedIn logo