IMF's Georgieva Warns AI Is a Slow-Moving Tsunami for Jobs

IMF Managing Director Kristalina Georgieva said on Monday that artificial intelligence is a slow-moving but unmistakable tsunami for global employment, and the IMF now expects about 40% of jobs worldwide to be augmented, downgraded or eliminated over time. She made the remarks as the IMF and World Bank annual meetings opened in Bangkok. Georgieva also said AI investment is acting as both a demand and an investment shock that could lift growth in advanced Asian economies by half a percentage point to a full point, and that it is helping hold global growth near 3% despite an energy supply shock.

Context — why the AI jobs warning lands now

The 40% global exposure figure builds on IMF research first published in January 2024, which put exposure at roughly 60% of jobs in advanced economies, 40% in emerging markets and 26% in low-income countries. That earlier work also found about half of exposed jobs in rich economies could benefit from AI rather than lose out, so the fund's own framing has always carried both an upside and a downside.

What makes Monday's version sharper is the sequencing. Georgieva said AI is already reducing demand for routine, repetitive work while adding lower-paid jobs in sectors such as construction, an outcome she said was nothing to celebrate. The hollowing is therefore visible before the headline job losses arrive.

Singapore President Tharman Shanmugaratnam, speaking at the same meetings, said the first rung of employment, the first job after graduation, is seeing less demand, with the clearest evidence in the sectors most exposed to AI. That gives the warning a concrete labour-market read rather than a forecast.

The macro backdrop matters because the same capital spending driving the displacement is also driving growth. AI investment is keeping global output near 3% even with an energy supply shock running through the system, which puts central banks in the awkward position of watching an investment boom that supports activity while it also pressures costs.

Data — what the numbers show

The IMF's exposure estimates span the income spectrum: roughly 60% of jobs in advanced economies, 40% in emerging markets and 26% in low-income countries. The current global figure Georgieva cited, about 40%, sits below the advanced-economy number because it is a worldwide aggregate.

The graduate data is the most concrete evidence of the shift. Around 74% of Singapore's 2025 university graduates in the labour force found full-time work, down from about 88% of the 2022 cohort, while the share unemployed and seeking work rose to about 10% from under 4%, according to a Ministry of Education survey.

Metric	2022 cohort	2025 cohort
Singapore graduates in full-time work	~88%	~74%
Unemployed and seeking work	under 4%	~10%

Academic work on the mechanism is not settled. A Harvard working paper found junior employment fell by about 9% at firms that adopted generative AI, relative to those that did not. A May 2026 study by researchers at the University of Warwick, the London School of Economics and Oxford found that much of AI's apparent effect on junior hiring faded once exposure to remote work was taken into account.

On the growth side, the IMF's estimate that AI investment could add half a point to a full point to advanced Asian economies is the number that matters for regional asset prices, and it is the one the fund is weighing against the energy shock in Tuesday's forecasts.

Analysis — what it means for markets and sectors

The growth estimate is the part of Georgieva's remarks with the most direct market transmission. Advanced Asian economies tied to the technology supply chain — Taiwan, South Korea and Singapore — are the named beneficiaries of that half-point to full-point lift, which supports regional equities and the currencies that trade alongside them.

The fiscal warning is the slower-burning risk. Georgieva said that by tilting economies toward capital and away from labour, AI could erode the taxes on work that have traditionally been governments' main source of revenue. For heavily indebted sovereigns, a shrinking labour tax base is a structural problem that arrives years before it shows up in a budget line.

The counter-argument is that the displacement evidence is softer than the headline suggests. The Warwick, LSE and Oxford study found much of the junior-hiring effect disappeared once remote-work exposure was controlled for, which means the graduate weakness in Singapore may not be purely an AI story.

Positioning follows the split. Money is going into the AI capital-spending complex on the growth argument while the labour-displacement trade remains a thematic rather than an index-level position, because no one has priced a tax-base erosion into sovereign curves yet.

Outlook — what to watch next

The IMF's World Economic Outlook, with updated growth forecasts, is due on Tuesday. That release is the next catalyst, and the question is whether the fund upgrades its numbers on the strength of AI investment or cuts them on the energy shock.

Beyond that, the labour data will do the arguing. Singapore's graduate employment survey is the closest thing to a live read on the first-rung effect, and further deterioration in the 74% full-time figure or the roughly 10% unemployment share would validate the displacement thesis faster than any IMF projection.

On the policy side, Georgieva urged policymakers to assess developments carefully and act on the lessons now. Any movement on how governments tax capital versus labour would be the first sign that the fiscal warning is being taken seriously.

Frequently Asked Questions

What does the IMF's 40% AI job exposure figure actually mean?

It means the fund expects about 40% of jobs worldwide to be augmented, downgraded or eliminated by AI over time, a global aggregate that sits below the roughly 60% exposure the IMF estimated for advanced economies in January 2024. The estimate covers a range of outcomes, not just job losses, because earlier IMF research found about half of exposed jobs in rich economies could benefit from AI.

Why are new graduates hit first by AI displacement?

Entry-level roles carry the routine, repetitive tasks that AI automates most easily, so employers cut demand at the first rung before restructuring senior roles. Singapore's data shows the effect: full-time employment for 2025 graduates fell to around 74% from about 88% for the 2022 cohort, with unemployment and job-seeking rising to roughly 10% from under 4%.

How does AI investment support growth while cutting jobs?

Georgieva described the boom as a demand and investment shock, meaning the capital spending itself creates activity even as it displaces labour in specific roles. The IMF estimates that spending could add half a percentage point to a full point to growth in advanced Asian economies, which is why global growth is holding near 3% despite an energy supply shock.

Bottom Line

The IMF is betting AI investment keeps global growth near 3% while it slowly erodes the labour tax base that funds sovereign budgets.

Disclaimer: This article is for informational purposes only and does not constitute investment advice. CFD trading carries high risk of capital loss.