AI and the job market: what the research says
The IMF, the World Economic Forum and Microsoft on how AI is changing work, and what it means for how you agree to be paid for project work.
Headlines about AI and jobs swing between "nothing changes" and "every job goes." Three widely cited sources give a more useful picture: an IMF analysis of which jobs are exposed, a World Economic Forum survey of employers, and Microsoft's research inside companies.
This post summarizes what each one says, what none of them say, and what it means in practice for people who contribute to companies.
What the IMF found
In January 2024 the International Monetary Fund estimated that almost 40% of global employment is exposed to AI, rising to about 60% in advanced economies. Its analysis split that exposure in two: roughly half of exposed jobs may benefit from AI, which could raise productivity, while in the other half AI may take over key tasks and lower demand for labor.
Exposure is not the same as job loss. It means AI can affect a meaningful share of the tasks in a job, for better or worse.
What employers told the World Economic Forum
The WEF's Future of Jobs Report 2025 surveyed more than 1,000 employers representing over 14 million workers. Those employers expect 170 million roles to be created and 92 million displaced by 2030, a net increase of 78 million. They also expect 39% of the key skills that jobs require to change over the same period.
These are employers' expectations, not a forecast for any individual career. But they point to a lot of movement: roles changing, new roles appearing and people switching between them.
What Microsoft sees inside companies
Microsoft's 2025 Work Trend Index, based on a survey of 31,000 workers in 31 countries plus LinkedIn and Microsoft 365 data, describes a new kind of "Frontier Firm" built around people working alongside AI agents. It reports that 45% of leaders say expanding team capacity with "digital labor" (AI agents) is a top priority for the next 12 to 18 months, second only to upskilling their existing workforce.
The report is written for large organizations, but the direction applies to small ones too: more of the routine work is done by software, and more of the value sits in judgment, relationships and knowing what to build.
What the research doesn't say
- It doesn't say every job disappears. The WEF numbers show more roles created than displaced.
- Exposure isn't displacement. The IMF counts jobs where AI touches key tasks, including jobs it may improve.
- The figures are averages. Global and national estimates say little about a particular person, company or industry.
- Projections change. Each source reflects surveys and assumptions from a specific point in time.
What it means in practice
If more work is organized around projects and small teams, more people will contribute to several companies instead of one, and the terms of each contribution matter more. Three posts pick up from here:
- Revenue share vs equity: how to decide what each contribution is paid in.
- Work should compound, not expire: why a small, capped share suits work that keeps earning.
- Work that keeps paying: how to keep track of shares across many projects.
For teams that form around a single launch or product, see AI will make pop-up companies normal.
Sources
- Kristalina Georgieva, IMF. AI will transform the global economy. Let's make sure it benefits humanity (opens in a new tab). January 14, 2024.
- World Economic Forum. The Future of Jobs Report 2025 (opens in a new tab). January 2025.
- Microsoft WorkLab. 2025: The year the Frontier Firm is born (opens in a new tab). April 2025.