Will AI create or take away more job opportunities?

Leans yes

AI is expected to create more job opportunities net than it eliminates over the medium to long term, while transforming most existing jobs and causing significant short- to medium-term disruption and displacement in specific roles. This aligns with historical patterns from previous general-purpose technologies (like computers or electricity), current empirical data showing limited mass job losses so far, and major forecasts from organizations like the World Economic Forum (WEF) and McKinsey.[1]

Historical Context and Current Evidence

Past technological shifts have typically been net job creators despite destroying some roles. For example, the rise of personal computing eliminated millions of jobs (e.g., typewriter-related or certain clerical work) but created far more in manufacturing, software, analytics, and related fields.[1]

As of 2025–2026 data:

•There is limited evidence of economy-wide AI-driven job losses or rising unemployment specifically in highly AI-exposed occupations since the widespread adoption of generative AI tools like ChatGPT (late 2022). Impacts appear more in slower hiring (especially for younger/entry-level workers in exposed fields) than in mass firings.[2]

•AI both substitutes for some tasks/workers and complements/augments others, often leading to productivity gains that can expand overall demand and employment at adopting firms. Some analyses show roughly offsetting effects (jobs removed via substitution vs. added via complementarity), with net effects modest so far and adjustment happening more through hiring patterns than layoffs.[3]

•Studies of firm-level AI investment often find accompanying employment growth (alongside sales growth), with shifts toward more skilled or educated workers. Productivity rises in AI-exposed areas, and specialist AI skills command wage premiums.[4]

•Early-career and certain white-collar/entry-level roles in exposed areas (e.g., some administrative, analytical, or creative tasks) show more vulnerability via reduced postings or slower growth. Overall labor markets have not shown an "AI jobs apocalypse."[5]

AI is primarily reshaping how work is done (task reallocation, hybrid human-AI workflows) rather than eliminating entire occupations outright in most cases.[1]

Key Projections

Major reports forecast net job creation amid disruption:

•WEF Future of Jobs Report 2025: Globally, by 2030, structural labor market shifts (heavily involving AI and related technologies) could create around 170 million jobs while eliminating about 92 million, for a net gain of roughly 78 million. AI and information processing are expected to transform businesses for 86% of surveyed employers and rank among the top job creators. Growing roles include AI/ML specialists, big data specialists, and fintech engineers; declining ones include certain clerical, administrative, and some creative/support positions. Many employers plan to hire for AI skills and retrain workers for human-AI collaboration.[6]

•McKinsey Global Institute (US-focused, to ~2035): Automation/AI could reduce labor demand equivalent to roughly 36 million jobs, but broader economic growth (including the AI value chain) and new demand could create about 40–41 million, resulting in more total jobs than today (net positive). However, around 11 million workers (roughly 7% of the workforce; range ~6–16 million) may need to switch to entirely different occupations—far above historical rates of occupational mobility (about 3.6x). Many more will see their current roles transformed (e.g., high task automation in office/administrative support). Growth areas include healthcare, construction, management, and AI-related roles; challenges center on workforce mobility and retraining rather than overall scarcity.[7]

•Other analyses (e.g., Goldman Sachs) note modest near-term net drags in some metrics offset partially by augmentation effects that boost employment in complementary roles, with younger/less-experienced workers more affected.[8]

New jobs emerge in AI development, deployment, maintenance, governance/ethics, infrastructure, human oversight of AI systems, and entirely new occupations enabled by higher productivity or novel applications. Augmented roles (where AI handles routine parts, freeing humans for judgment, creativity, interpersonal skills, or complex problem-solving) can also expand.

Important Caveats and Challenges

•Displacement and inequality risks are real: Certain routine cognitive/administrative, some creative, and entry-level tasks face higher substitution risk. Lower-wage or less-adaptable workers may struggle more with transitions. Without strong reskilling, education reform, and support, inequality could widen, and some groups (e.g., young workers entering the market) face tighter conditions.[9]

•Transformation > pure elimination: Most jobs will change substantially (tasks reshuffled, AI collaboration required). Success depends on redesigning workflows, not just bolting AI onto old processes.[1]

•Timing and uncertainty: Near-term effects are milder and more about hiring/composition shifts. Longer-term outcomes depend on adoption speed, complementary investments (skills, infrastructure, energy), policy (education, safety nets, transition support), and whether productivity gains translate into broad economic expansion and new demand. Extreme scenarios (very rapid AGI-like advances) remain speculative and outside most baseline forecasts.

•Outcomes are not predetermined—they depend on choices by businesses, governments, educators, and workers around training, AI deployment ethics, and inclusive growth.[10]

In summary, evidence and leading forecasts indicate AI will likely be a net creator of job opportunities (as with prior technologies), driven by productivity, new industries/roles, and expanded demand, but it will take away or heavily alter many existing ones in the process. The bigger story is rapid change requiring adaptation: workers who build AI fluency, complementary human skills (creativity, critical thinking, emotional intelligence, physical/hands-on expertise where relevant), and lifelong learning will be best positioned. Policymakers and employers play key roles in facilitating smooth transitions to realize the upside while mitigating downsides.

Citations (10)

1mit.eduAcademic
https://mitsloan.mit.edu/ideas-made-to-matter/ai-and-jobs-4-takeaways-googles-chief-economist
Backs: “This aligns with historical patterns from previous general-purpose technologies (like computers or electricity), current empirical data showing limited mass job losses so far, and major forecasts from organizations like the World Economic Forum (WEF) and McKinsey.”
evidence, article, points to yes. Claim: Past technologies like personal computing were net job creators, and AI mainly reshapes how work is done rather than eliminating occupations.
2naraway.comLow provenance
https://naraway.com/Blogs/ai-jobs-hiring-evidence-2026-employers-guide.html
Backs: “Impacts appear more in slower hiring (especially for younger/entry-level workers in exposed fields) than in mass firings.”
evidence, article, points to yes. Claim: Limited evidence of economy-wide AI job losses; impact appears as slower hiring for entry-level workers rather than mass firings.
3cepr.orgNonprofits and think tanks
https://cepr.org/voxeu/columns/hiding-mean-ai-adding-jobs-and-removing-them-same-time
Backs: “Some analyses show roughly offsetting effects (jobs removed via substitution vs. added via complementarity), with net effects modest so far and adjustment happening more through hiring patterns than layoffs.”
evidence, article, neutral. Claim: AI substitutes some tasks and complements others, with roughly offsetting effects and modest net effects so far.
4brookings.eduNonprofits and think tanks
https://www.brookings.edu/articles/the-effects-of-ai-on-firms-and-workers/
Backs: “Productivity rises in AI-exposed areas, and specialist AI skills command wage premiums.”
evidence, article, points to yes. Claim: Firm-level AI investment is often accompanied by employment growth, skill shifts, productivity gains and wage premiums for AI skills.
5theguardian.comNews and mediaquality 0.75
https://www.theguardian.com/technology/2026/aug/12/ai-job-destruction
Backs: “Overall labor markets have not shown an "AI jobs apocalypse."”
evidence, article, neutral. Claim: Entry-level and white-collar roles are more vulnerable via reduced postings, but there has been no AI jobs apocalypse.
6arstechnica.comNews and mediaquality 0.95
https://arstechnica.com/ai/2025/01/ai-could-create-78-million-more-jobs-than-it-eliminates-by-2030-report/
Backs: “Many employers plan to hire for AI skills and retrain workers for human-AI collaboration.”
statistic, article, points to yes. Claim: WEF 2025 forecasts 170 million jobs created and 92 million eliminated by 2030, a net gain of about 78 million.
7yahoo.comNews and mediaquality 0.69
https://finance.yahoo.com/technology/ai/articles/mckinsey-research-shows-ai-creating-110000707.html
Backs: “Growth areas include healthcare, construction, management, and AI-related roles; challenges center on workforce mobility and retraining rather than overall scarcity.”
statistic, article, points to yes. Claim: McKinsey projects about 36 million jobs reduced versus 40-41 million created in the US, with 11 million workers needing to switch occupations.
8goldmansachs.comCompanies
https://www.goldmansachs.com/insights/articles/the-jobs-ai-is-likely-to-boost-and-those-it-may-disrupt
Backs: “Other analyses (e.g., Goldman Sachs) note modest near-term net drags in some metrics offset partially by augmentation effects that boost employment in complementary roles, with younger/less-experienced workers more affected.”
evidence, article, neutral. Claim: Goldman Sachs notes modest near-term net drags partly offset by augmentation, with younger workers more affected.
9fortune.comNews and mediaquality 0.88
https://fortune.com/2026/10/03/ai-job-displacement-creation-new-careers-11-million-americans-workforce-mobility/
Backs: “Without strong reskilling, education reform, and support, inequality could widen, and some groups (e.g., young workers entering the market) face tighter conditions.”
evidence, article, points to no. Claim: Displacement and inequality risks are real, especially for lower-wage workers and young entrants, without strong reskilling.
10weforum.orgNonprofits and think tanks
https://www.weforum.org/stories/jobs-and-the-future-of-work/ai-work-jobs-intelligence-era/
Backs: “Outcomes are not predetermined—they depend on choices by businesses, governments, educators, and workers around training, AI deployment ethics, and inclusive growth.”
attribution, article, neutral. Claim: Outcomes depend on choices by businesses, governments, educators and workers around training and deployment.