Will AI create or take away more job opportunities?

Will AI be a net job creator or eliminator? I'll pull the latest forecasts on displaced and emerging roles.

The short answer is: AI will do both - and most large-scale forecasts expect it to create slightly more jobs than it takes away in aggregate by 2030, but with massive churn and uneven effects.

That is not a consensus that no one loses their job - it is a forecast of net gain alongside large displacement.

1. What the most-cited forecasts actually say

The most comprehensive global employer survey is the World Economic Forum's Future of Jobs Report 2025:

•It surveyed 1,000+ employers across 55 economies

•It projects 92 million jobs displaced, but 170 million created — a net gain of 78 million

•Other summaries of the same report phrase it as : The World Economic Forum (WEF) predicts that AI will displace 92 million jobs worldwide by 2030, but it will also create 170 million new jobs, resulting in a net gain of 78 million jobs.

•In more detail: from 2025 to 2030 employers expect both job creation and job displacement, with 170 million new jobs created and 92 million displaced, producing a net gain of 78 million jobs globally

In other words, employers expect a structural labour market churn of about 22% of current jobs, but net employment still +7%.

Other institutions frame it differently but point in the same direction for the medium term:

•Gartner's 2025 AI Job Impacts Analysis found that starting in 2028-2029, AI will create more jobs than it eliminates

•An earlier WEF headline often cited was more conservative: AI and information-processing technologies to create 11 million jobs and displace 9 million by 2030 - showing how much estimates vary by definition and time horizon.

2. Why "exposed" does not mean "eliminated"

Much confusion comes from mixing exposure with displacement:

•The IMF says nearly 40 percent of global employment is exposed to AI, rising to 60 percent in advanced economies

•Goldman Sachs said in 2023 that AI could substitute for up to a quarter of current work, exposing 300 million jobs in the US and Europe

Goldman itself clarifies the nuance: "exposed" doesn't mean displaced. Goldman estimates 7% of US workers will actually lose their jobs, while 63% are augmented. The actual prediction: 25% of work tasks are automatable by generative AI, but only 7% of workers will be fully displaced. 63% will be augmented — doing the same job with AI handling some tasks.

McKinsey makes a similar distinction:

•57% of US work hours automatable with current technology is technical potential now, not a forecast of job losses. It calls it technical potential, "not a forecast of job losses" in other summaries.

•About 50% of work activities automatable globally is the theoretical ceiling, while the near-term transition need is estimated at 12 million US workers needing occupational transitions by 2030.

3. Which jobs are most affected

Most likely to be displaced or sharply reduced (where AI automates tasks):
WEF's fastest-declining roles list includes Data entry clerks, Administrative assistants, Bookkeepers & payroll clerks, Bank tellers & cashiers, Postal service clerks - generally routine clerical, data-processing and customer-service work. 41% of employers surveyed are planning AI-related headcount reductions, and 40% anticipate cuts where AI automates tasks.

Most likely to be created or to grow (where AI complements or creates new needs):

•AI & machine learning specialists, Data analysts & scientists, Big data specialists, Security analysts, Renewable energy technicians are the WEF's fastest-growing role categories.

•Other growth clusters cited across reports: AI system management, data engineering, cybersecurity, green technology, and human-AI collaboration roles that did not exist a decade ago.

The pattern is consistent with history: 60% of today's jobs didn't exist in 1940 — new technology creates new work categories like AI prompt engineers, AI safety researchers, and autonomous systems supervisors.

4. The catch: net gain does not mean painless

1.Timing gap: Displacement can happen faster than creation. Freelance writing, coding, translation and entry-level white-collar work are already seeing contraction, while new AI-adjacent roles require fundamentally different skills.

2.Geography and inequality: Advanced economies face 60% AI exposure vs 26% for low-income countries, but advanced economies also have stronger safety nets and capacity to create new AI-adjacent roles. Middle-income countries with high exposure but weaker adaptation systems face higher risk.

3.Augmentation vs automation: Jobs where AI complements the worker (e.g., doctors using AI diagnosis, designers using generative tools) see productivity and wage gains. Jobs where AI substitutes core tasks see displacement risk. About half of AI-exposed jobs in advanced economies will actually benefit from AI, the other half face displacement risk according to the IMF framework.

> Andrew Ng's summary often quoted: "AI will not replace you. A person using AI will."

Bottom line

If you look at headline numbers through 2030, credible institutional forecasts — WEF, Goldman Sachs, IMF, McKinsey — all project net positive job creation (+78 million net in the WEF baseline), not mass unemployment.

If you look at the individual level, AI will take away more opportunities in routine, predictable tasks and create more opportunities in technical, creative and human-centric tasks that work with AI.

Whether that net gain feels like gain or loss for any specific worker depends almost entirely on reskilling, mobility, and how quickly education and policy catch up to the churn.

Citations (1)

1jobzonerisk.comData and tools
AI Job Loss Predictions [October 2026]
Backs: “Goldman Sachs said in 2023 that AI could substitute for up to a quarter of current work, exposing 300 million jobs in the US and Europe. Goldman itself clarifies the nuance: "exposed" doesn't mean displaced.”