Skip to content

Essentials: jobs

Will AI destroy more jobs than it creates? The key papers, numbers, people and newsletters, sized to the time you have. The full question page. Picks are chosen for their weight in the debate and span its sides; they are refreshed every two weeks, last on 29 Oct 2026. The reasons are AI-written and evidence-cited; no rankings or scores.

Time

Five papers, the numbers with their exact quotes, five people and three sources.

Papers to read

  • The future of employment: How susceptible are jobs to computerisation?Carl Benedikt Frey and Michael A. Osborne · Technological Forecasting and Social Change · 2016

    The 2013 estimate, published in 2016, that 47% of US jobs are at high risk of computerisation, still quoted in headlines.

    Frey and Osborne estimate how likely 702 US occupations are to be computerised, using expert judgements of which tasks machines can do. They classify about 47% of US employment as at high risk of automation over an unspecified number of years, a widely cited estimate of exposure rather than of jobs actually lost.

    Full text at Technological Forecasting and Social Change

  • GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language ModelsTyna Eloundou et al. · arXiv · 2023

    Measures exposure to language models task by task: 80% of workers have some tasks exposed, 19% have half or more.

    Combining human ratings with GPT-4 classifications, finds that roughly 80% of the U.S. workforce could see at least 10% of their tasks affected by large language models, and about 19% could see over half their tasks affected, with higher-wage occupations facing greater exposure than lower-wage ones. With an LLM alone about 15% of all worker tasks could be completed significantly faster, rising to 47-56% when LLM-powered software tools are included — evidence often cited to argue AI functions as a general-purpose technology with economy-wide reach rather than a narrow labor-saving tool.

    Full text at arXiv

  • Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial IntelligenceErik Brynjolfsson et al. · Stanford Digital Economy Lab working paper · 2025

    Payroll data showing a 13% relative fall in employment for 22 to 25 year olds in the most AI-exposed jobs.

    Using payroll data, the authors find that since generative AI spread, workers aged 22-25 in the most AI-exposed occupations saw a 13% relative decline in employment, even after controlling for firm-level shocks.

    Full text at Stanford Digital Economy Lab working paper

  • The Future of Jobs Report 2025World Economic Forum · World Economic Forum · 2025 · Evidence against the claim

    Employer survey expecting 8% of current jobs displaced by 2030 but net employment growth of 7%.

    Full text at World Economic Forum

  • Forecasting the Economic Effects of AIEzra Karger et al. · NBER · 2026 · Evidence for the claim

    A 2026 survey of five groups on jobs and growth: small declines in participation by default, larger under rapid AI progress.

    The authors survey five groups (academic economists, AI company staff, AI policy researchers, top forecasters and the public) on how AI will affect the US economy. The median respondent in each group expects annual GDP growth of 2.5% and small declines in labor force participation; under a rapid-progress scenario, experts expect growth near 4% and participation falling to 55% by 2050, about half of it attributed to AI.

    Full text at NBER

Of these 3 papers: 3 not coded on a side.

Of these 5 papers: 1 with evidence for the claim, 1 with evidence against, 3 not coded on a side.

Of these 5 papers: 1 with evidence for the claim, 1 with evidence against, 3 not coded on a side.

Figures to know

  • 47%US jobs at high risk of computerisation
    Carl Benedikt Frey and Michael A. Osborne, 2016 · Jobs
    “According to our estimate, 47 percent of total US employment is in the high-risk category”
  • 9%Jobs at high risk of automation, average of 21 OECD countries
    Melanie Arntz et al., 2016 · Jobs
    “Overall, we find that, on average across the 21 OECD countries, 9 % of jobs are automatable.”
  • 80%Workers with at least 10% of tasks affected by LLMs
    Tyna Eloundou et al., 2023 · Jobs
    “around 80% of the U.S. workforce could have at least 10% of their work tasks affected by the introduction of LLMs”
  • −13%Employment of early-career workers (22-25) in the most AI-exposed jobs, relative
    Erik Brynjolfsson et al., 2025 · Jobs
    “early-career workers (ages 22-25) in the most AI-exposed occupations have experienced a 13 percent relative decline in employment”
  • 6.5% (range 6 to 7)US workforce displaced if AI is widely adopted
    Goldman Sachs Research, 2025 · Jobs
    “could displace 6-7% of the US workforce if AI is widely adopted”

Every number is quoted from its source and linked to the paper; estimates differ in method and horizon, so they are not directly comparable.

People to follow

Of these 3 people: 1 leans agree, 2 leans disagree. Documented, cited positions only.

Of these 5 people: 1 agrees, 1 leans agree, 1 it's complicated, 2 leans disagree. Documented, cited positions only.

Of these 5 people: 1 agrees, 1 leans agree, 1 it's complicated, 2 leans disagree. Documented, cited positions only.

Newsletters and sources