Will AI create or take away more job opportunities?

This is a widely debated question among economists, researchers, and policymakers, with legitimate arguments on both sides. Let me look into recent analyses and data on this topic.This is genuinely one of the most contested questions in economics right now, and the honest answer is: it depends on the timeframe, and experts disagree even on the data we have today. Here's what the evidence actually shows:

The Case for "More Jobs Overall" (Long-term, Optimistic View)

The World Economic Forum's Future of Jobs Report 2025 is the most-cited source for net-positive predictions, projecting that job disruption will equate to 22% of jobs by 2030, with 170 million new roles set to be created and 92 million displaced, resulting in a net increase of 78 million jobs. However, it's worth noting a methodological nuance: the report's own figure specifically for AI and information processing is 11 million jobs created and 9 million displaced — the larger 170M/92M figures combine all trends (AI, demographics, green transition, trade tensions, economic uncertainty), not AI alone.

McKinsey's recent analysis reaches a similar "net positive but painful" conclusion: artificial intelligence could create more jobs than it destroys, but millions of Americans could still be left scrambling, with anywhere from 6 million to 16 million workers potentially needing to switch occupations. Specifically, roughly 11 million American workers, or about 7% of the current workforce, may need to change jobs by 2035 because of AI-related displacement.

The Case for "No Net Effect Yet" (Current Reality)

Several 2026 studies suggest we haven't seen large-scale disruption yet. An NBER survey found that AI has had zero net effect on firm employment over the last three years, and a majority predict little net impact over the next three years. Similarly, the Yale Budget Lab's ongoing monitoring found that where AI is used most, there is neither higher unemployment nor lower employment, and the types of jobs offered by the market are not changing in a way that can be attributed to AI.

Research on augmentation vs. automation also supports this: the Anthropic Economic Index shows that augmentation — where users learn, iterate or validate with AI — has surpassed pure automation, with one 2026 preprint estimating 78.7% of observed AI interactions are augmentation rather than automation.

The Case for Real, Immediate Disruption

Other data paints a more worrying short-term picture. AI was the leading stated reason for US layoffs in 2026, with 116,175 cuts through August. The effects are hitting young and entry-level workers particularly hard: entry-level hiring plunged 73.4% and admin role hiring dropped 35.5% in 2025, with half of employers citing AI automation as the cause, while young workers' employment share in AI-exposed occupations fell from 16.4% to 15.5% since ChatGPT's launch.

Goldman Sachs' economist offers a middle estimate: the base case is that 6–7% of US workers will be displaced over a roughly 10-year adoption period, adding about 0.6 percentage points to the unemployment rate — though if the transition is frontloaded, disruption could be significantly larger.

The Key Problem Isn't Totals — It's Mismatch

Perhaps the most important insight across studies is that the aggregate numbers may matter less than who loses jobs versus who gets new ones. McKinsey found only one in seven displaced workers has a direct path to a growing job, and nearly half face what the report calls an "unpaved" path, blocked by large skill gaps or credential requirements. A bookkeeper being automated out of a job can't simply become a machine learning engineer overnight — geography, education, and retraining time create real hardship even in a "net positive" scenario.

There's also a wage-divide emerging among those who stay employed: AI-skilled workers now earn 62% more than non-AI peers in equivalent roles — up from 56% a year ago and 25% two years ago, with workers who've developed AI expertise commanding historically large and widening wage premiums.

Bottom Line

Most major institutional forecasts (WEF, McKinsey, Goldman Sachs) lean toward net job creation over the long run (5-10+ years), following the historical pattern of technologies like the printing press, electrification, and computerization. But in the short-to-medium term, there's real, measurable disruption concentrated in entry-level, administrative, and routine white-collar work — and the transition will likely be bumpy, uneven, and genuinely painful for specific groups of workers, even if the aggregate numbers eventually turn positive. As one analysis candidly put it, the honest answer is that we do not know yet, but there are signals forming on both sides.