GCRI · Future of Work and Skills
Automation and jobs
Automation, robotics and AI take over some tasks, assist others and create new ones. Whether this displaces workers, raises productivity and wages or does both depends on the tasks involved, how firms deploy technology and how institutions support transitions. This page sets out how automation affects jobs, how exposure and impact are measured, and what evidence serves the policymakers, employers and unions responsible.
Automation and jobs on the record.
Automation and jobs: how the risk develops.
AI/robotics impact and job redesign. The dossier below sets out how the risk develops, what is measured, where the data come from, how it is modelled and who decides.
Tasks, not whole jobs
Most occupations combine tasks that can be automated with tasks that cannot. Automation reshapes jobs more often than it eliminates them outright.
Displacement and reinstatement
Technology displaces labour from some tasks and creates new tasks that need people. Net effects depend on the balance between the two.
Productivity gains may not reach wages
If gains accrue mainly to capital owners, automation can raise output while wages stagnate for exposed workers.
Transitions are costly for individuals
Displaced workers often face long unemployment or lower pay in new jobs, especially older workers and those in declining regions.
How automation and jobs is modelled, measured and tested.
The model, its assumptions, its data and its uncertainty, in the form a scientist or engineer would check. Structures follow published methods. The values drawn are illustrative, not findings.
System model
The couplings between the systems in automation and jobs: signs of influence, a reinforcing and a balancing loop, and cross-links where stress jumps.
Automation exposure differs by domain.
Findings are designed for the Exponential Technology series. The ILO’s 2023 working paper estimates exposure, which is not job loss.
- Firm adoption data.
- Worker surveys.
What is measured, and how well.
Each indicator carries its unit, its cadence, the kind of source behind it and the evidence label a value would carry. No values are shown: a programme supplies them under its own authority.
| Indicator | Unit | Cadence | Source type | Label |
|---|---|---|---|---|
| Employment in occupations with high automation exposure | percent of employment | per assessment | Task-based exposure studies | ESTIMATED |
| Firms using AI or robots | percent of firms | per survey | Business surveys | REPORTED |
| Displaced worker re-employment rate | percent re-employed within a year | per survey | Displaced worker surveys | REPORTED |
| Wage change after displacement | percent | per survey | Linked employer-employee data | ESTIMATED |
| Labour share of income | percent of GDP | annual | National accounts | ESTIMATED |
| New job titles emerging | count | annual | Occupational classification updates and job postings | OBSERVED |
Where the evidence comes from.
Named public datasets and institutional records, each with its licence class. Sources by agreement are used only under their terms and never republished.
| Source | What it gives | Licence class |
|---|---|---|
| ILO global studies on generative AI exposure | Occupational exposure estimates by country and gender | Public, ILO terms |
| OECD studies on automation and AI in the labour market | Evidence on exposure and adoption | Public, OECD terms |
| US Bureau of Labor Statistics Displaced Worker Survey | Outcomes of displaced workers | Public domain, US government |
| O*NET occupational task database | Task content of occupations used in exposure studies | Open, CC BY 4.0 |
| Eurostat enterprise AI and robot use statistics | Adoption by firms | Open, Eurostat reuse policy |
Who decides, and what the evidence serves.
Future of Work and Skills prepares evidence. The decisions below are taken by the authorities named, under their own rules, and never by the platform.
| Who decides | The question | Evidence this page serves |
|---|---|---|
| Labour ministries | Which workers and regions need transition support | Exposure profiles; displacement evidence |
| Employers and unions | How to introduce technology and share gains | Adoption and wage evidence |
| Training providers | Which skills to build for new tasks | Emerging task evidence |
Outputs this topic produces
- Task-based exposure profiles by occupation and group
- Displacement and re-employment analyses
- Evaluation designs for transition programmes
- Evidence notes for social dialogue on technology
What researchers ask about automation.
- Which tasks can current AI and robots perform reliably and at acceptable cost?
- When does technology substitute for workers and when does it complement them?
- How are productivity gains shared between firms, workers and consumers?
- How can jobs be redesigned so people keep meaningful, skilled work?
Displacement, productivity and new work.
Automation has a displacement effect, as machines take over tasks, and a productivity effect, as lower costs raise demand and employment elsewhere. Historically, new tasks and occupations have also emerged, reinstating demand for labour. The balance between these forces determines the net effect on employment and wages.
Industrial robots mainly automate physical, repetitive tasks in manufacturing and logistics. Generative AI reaches cognitive tasks such as drafting, summarising and coding. Evidence from early workplace studies suggests AI assistance can raise productivity, sometimes most for less experienced workers, but effects vary widely by task and implementation. What employers choose to automate, and how, matters as much as technical capability.
Evidence on automation and its limits.
- 01
Task exposure measures
Matching technology capabilities to occupational task descriptions, informed by artificial intelligence and machine learning and robotics and automation. Measures depend on judgements about capability that age quickly.
- 02
Robot adoption studies
Linking robot installations to local employment and wages with econometrics. Results differ between countries and periods.
- 03
Workplace experiments
Randomised or natural experiments in firms adopting AI, using social and behavioural science. Short-run effects may not reflect long-run change.
- 04
Job redesign case studies
How organisations reorganise work around new tools. Case studies illustrate possibilities rather than averages.
How it is studied, and what the methods cannot say.
- 01Economics and development economicsTask-based analysis of technology · Labour market institutions · Just transition economicsApplied here
- 02Artificial intelligence and machine learningSkills extraction from job postings · Job matching in employment servicesApplied here
- 03Robotics and automationRobot adoption and employment · Safety in human-robot work · Job redesign around robotsApplied here
- 04EconometricsDisplacement effects with linked data · Training programme evaluation · Workforce projectionsApplied here
Where this page’s references lead.
The bodies and documents this page is read against. Status and editions change; each link goes to the issuing body.
Where this page leads across the Consortium.
Questions and answers.
Will automation cause mass unemployment?
What is task-based analysis?
Who is most exposed to automation?
Is automation exposure the same as job loss?
Does AI help or hurt less experienced workers?
Can job redesign prevent displacement?
Does the platform advise firms on automation?
Contribute evidence from this page.
Reports is the Consortium’s quarterly publication. What you know about this subject can become evidence, a case or a correction in the next edition, credited to you and published under CC BY 4.0 where it is your own work.
- EmailGateway GW-GCRISend it from this pageEvidence, a correction, a case, a method, a review, a translation or profile input. It arrives tagged with this page, so the editor who owns it receives it.
- ZenodoOpen depositDeposit it and submit to the communityPublish your own record on Zenodo with your ORCID iD, then submit it to the Reports community there. Curators review it; the DOI stays yours.
- ReportsGCRISee every way to contributeThe contribution types, how each is reviewed, what credit you receive, the licence and what is never accepted.
Never send personal data, sensitive infrastructure detail or anything you have no right to share. A contribution is reviewed before use; sending it confers no authorship or editorial rights, and sponsorship never buys a finding.
Bring a method, a dataset or a question.
Information and enquiries are open today. An enquiry receives a reference and names the next step; applications, payments, compute allocations and registrations open only when their team, terms and review capacity exist.
Legal documents.
Maps, data and figures on this site are shown for context and general information. They are not advice, forecasts or warnings; the limits of the information are set out in the Disclaimers and Data Notices.