Goldman Sachs Says AI Is Putting Pressure on Global Jobs

Key Highlights of the News

• Goldman Sachs says AI exposure is already associated with weaker job-opening and employment growth in several developed economies.
• Call centres, software publishing, management consulting and advertising are among the industries showing employment weakness below historical trends.
• Entry-level workers appear more exposed, with AI-related hiring pressures more pronounced in junior roles.
• AI adoption across major developed economies is estimated at roughly 15%–20%, suggesting labour-market effects could become more visible as adoption expands.

As AI changes how organisations allocate work and hire talent, understanding AI technologies is becoming increasingly relevant across roles. Explore upGrad's Artificial Intelligence courses to build practical knowledge of AI, machine learning and emerging applications.

Goldman Sachs Finds AI Pressure Emerging Across Developed Economies

Artificial intelligence is beginning to affect labour markets in developed economies, with industries more exposed to AI automation showing weaker employment and job-opening growth, according to research cited by Goldman Sachs. The effects remain uneven, but the pressure is particularly visible in sectors where AI tools can automate parts of existing workflows.

Goldman Sachs found that industries with greater exposure to AI automation have experienced slower growth in job openings since the second half of 2022. The relationship has been particularly evident in countries including the US, Germany and Australia.

Information and communication services, one of the sectors with relatively high AI exposure, have also recorded slower employment growth across major developed economies since 2022. Employment in call centres, software publishing, management consulting and advertising has fallen below historical trends, according to the research.

The call-centre data illustrates the scale of the shift in some markets. Benzinga, citing the research reported by CNBC, said call-centre employment was 39% below trend in the US, 33% below trend in Canada and 27% below trend in Germany.

Entry-Level Workers Face Greater AI-Related Hiring Pressure

The impact is not distributed evenly across workers. Goldman Sachs analysed more than 800 occupations and found that AI-related headwinds were particularly pronounced among people entering the workforce. A 10% occupational exposure to AI was associated with a relatively small drag on annual headcount growth in France, Canada and the US, but the effect was larger for entry-level workers in countries such as Australia and the US.

The development is consistent with earlier research showing that AI exposure can affect younger workers before it produces broad-based unemployment. Goldman Sachs has previously estimated that roughly 6%–7% of workers could eventually be displaced in the US under its base-case scenario for widespread AI adoption, although the timing of that transition remains uncertain.

At the same time, the latest findings do not point to a broad collapse in employment. Other analysis cited by Benzinga suggests that highly AI-exposed industries have largely moved sideways rather than experiencing a wholesale employment decline, while less-exposed industries have continued to grow. The broader pattern indicates that AI may initially change tasks, hiring patterns and the composition of teams rather than eliminate entire occupations.

AI Adoption Could Intensify the Labour-Market Shift

The findings come as AI adoption continues to expand. Goldman Sachs estimates that adoption across major developed markets is currently around 15%–20%, with France, the US, the Netherlands and the UK among the countries reporting higher adoption levels. Major emerging markets have estimated adoption rates of roughly 10%–15%.

That expansion could make workforce restructuring more visible, particularly in jobs involving repetitive information processing, customer support, software development and other activities where AI systems can automate individual tasks.

However, the employment impact is unlikely to be uniform. Goldman Sachs CEO David Solomon has argued that while AI can eliminate some jobs, it can also create demand for other roles. He has pointed to emerging work around managing agentic AI systems, implementing workflows, compliance and validation, where human judgement remains important.

The immediate picture, therefore, is less about AI replacing the entire workforce and more about how organisations redistribute tasks between people and machines. For workers, particularly those entering exposed professions, the shift could increase the importance of AI literacy and specialised skills as employers adapt to increasingly AI-enabled workplaces.

Conclusion

Goldman Sachs' latest findings suggest that AI is no longer only a future consideration for labour markets. Employment and hiring patterns in several AI-exposed industries are already showing signs of pressure, with entry-level workers facing a particularly difficult transition.

The data does not yet indicate economy-wide AI unemployment. But as adoption increases, the balance between human work, automation and newly created AI-related roles is likely to become a more important factor in workforce planning and career decisions.

As organisations adopt AI more widely, they also need professionals who understand both intelligent systems and the security risks surrounding them. The IIT Delhi Cyber Security and AI Certificate Programme is relevant for professionals looking to build knowledge across AI and cybersecurity as these technologies increasingly converge.