Will AI create or take away more job opportunities?
AI is expected to create more job opportunities than it eliminates overall (a net positive), but it will displace millions of roles/tasks and require large-scale worker transitions, reskilling, and adaptation. The outcome is not predetermined—it depends heavily on the pace of adoption, how businesses deploy AI (for pure cost-cutting vs. growth and augmentation), education/training systems, and public policy.[1]
Historical Context and Current Evidence
Past technological waves (mechanization of agriculture, industrial automation, computers, the internet) destroyed specific jobs while creating more new ones in expanding sectors, raising productivity, living standards, and overall employment over time. AI follows a similar pattern but can act faster on cognitive and knowledge-work tasks.[2]
So far (through ~2025–2026 data), there is no evidence of mass unemployment from AI:
Crude “exposure” estimates (e.g., high percentages of tasks automatable) overstate job losses because jobs consist of many tasks, firms grow with productivity, and adoption takes time.[4]
Major Forecasts: Net Creation with Major Transitions
Leading analyses project net job gains, alongside significant churn:
Other research (Goldman Sachs, academic studies, OECD/EU analyses) similarly finds simultaneous adding and removing of jobs, with net effects depending on measurement of substitution vs. complementarity. Productivity gains from AI can lower costs, raise demand, and spawn new products/industries.[8]
Jobs/roles likely boosted: AI specialists, data scientists/engineers, roles involving human-AI collaboration, healthcare/care work, education, green energy/construction, creative/oversight/ethics roles, and those requiring complex interpersonal skills, judgment, or physical dexterity hard to automate. Augmentation can expand employment even in somewhat exposed fields.[5]
Jobs/roles more at risk of disruption: Routine administrative, clerical, customer service, some data entry/analysis, certain entry-level white-collar/coding, insurance claims, and repetitive cognitive tasks. Lower-wage or less-experienced workers can face higher substitution risk in some models; high-skill complex roles may be more exposed to tools but less fully replaceable.[9]
Key Caveats and Risks
Bottom Line and What Helps
AI will transform far more jobs than it fully eliminates, shifting tasks and creating demand for new skills (AI literacy + human strengths like creativity, critical thinking, resilience, collaboration, and domain expertise). Net job opportunities should rise with productivity and economic growth, consistent with history and current forecasts from McKinsey, WEF, and others. However, without proactive measures—lifelong learning, accessible/stackable credentials, apprenticeships, skills-based hiring (reducing unnecessary degree barriers), targeted support for displaced workers, and policies encouraging AI for augmentation/growth—the human costs of transition could be high for many individuals and communities.[12]
The future of work will increasingly involve human-AI partnerships. Societies and individuals that invest in adaptability are positioned to capture more of the upside.
Citations (12)
- https://finance.yahoo.com/technology/ai/articles/ai-create-more-jobs-kills-080000226.html
Backs: “The outcome is not predetermined—it depends heavily on the pace of adoption, how businesses deploy AI (for pure cost-cutting vs. growth and augmentation), education/training systems, and public policy.”
- https://news.mit.edu/2026/technology-creates-jobs-young-skilled-workers-ai-0521
Backs: “AI follows a similar pattern but can act faster on cognitive and knowledge-work tasks.”
- https://aisafe.careers/layoffs
Backs: “AI-attributed layoffs, while rising as a cited reason in announcements (e.g., tens to over 100k in periods of 2026 per Challenger data), remain a small fraction of total layoffs and discharges.”
- https://www.rand.org/pubs/commentary/2025/10/ai-is-making-jobs-not-taking-them.html
Backs: “AI-exposed occupations have often seen employment growth overall (not shrinkage), though with nuances by age, gender, and experience.”
- https://www.goldmansachs.com/insights/articles/the-jobs-ai-is-likely-to-boost-and-those-it-may-disrupt
Backs: “Net effects to date have been modest, sometimes a small drag on payroll growth, concentrated more on younger/less-experienced workers in exposed fields (e.g., entry-level white-collar or certain tech roles showing relative declines).”
- https://www.brookings.edu/articles/still-waters-rapid-currents-early-labor-market-transformation-under-generative-ai/
Backs: “Earnings and hours have often stayed stable even as adoption rises.”
- https://www.weforum.org/stories/jobs-and-the-future-of-work/future-of-jobs-report-2025-jobs-of-the-future-and-the-skills-you-need-to-get-them
Backs: “Employers plan heavy upskilling (majority) but some workforce reductions.”
- https://cepr.org/voxeu/columns/hiding-mean-ai-adding-jobs-and-removing-them-same-time
Backs: “Productivity gains from AI can lower costs, raise demand, and spawn new products/industries.”
- https://www.bis.org/publ/work1207.htm
Backs: “Lower-wage or less-experienced workers can face higher substitution risk in some models; high-skill complex roles may be more exposed to tools but less fully replaceable.”
- https://www.pewresearch.org/short-reads/2026/10/22/men-are-experiencing-more-job-growth-than-women-in-ai-exposed-occupations/
Backs: “Impacts can be unequal (e.g., by age—younger workers hit first in some data; gender differences in growth; regional variations).”
- https://www.weforum.org/stories/economic-growth/four-ways-ai-impact-job-markets
Backs: “Some scenarios include larger disruptions if adaptation lags.”
- https://www.mckinsey.com/mgi/media-center/mckinsey-research-shows-ai-creating-more-jobs-than-automation-displaces-but-11-million-american-workers-will-need-help-this-decade
Backs: “However, without proactive measures—lifelong learning, accessible/stackable credentials, apprenticeships, skills-based hiring (reducing unnecessary degree barriers), targeted support for displaced workers, and policies encouraging AI for augmentation/growth—the human costs of transition could be high for many…”