Automation could affect 36 million jobs by 2035, according to a recent report from the McKinsey Global Institute, raising fresh questions about worker training and economic policy.
The global research organization’s estimate points to a major shift over the next decade. Employers may assign more routine tasks to software, machines, and artificial intelligence systems. The forecast does not state that every affected job will disappear. Instead, many roles may change as technology handles a greater share of daily work.
What the Estimate Means
The figure covers jobs that could be affected, a broader measure than jobs expected to be eliminated. An affected worker might keep the same title while taking on new duties. Another worker may move into a related role after receiving training.
“Automation could affect 36 million jobs by 2035,” the McKinsey Global Institute reported.
The distinction matters for businesses and policymakers. A forecast about exposure to automation does not, by itself, show how many people will lose work. The final outcome will depend on adoption rates, investment costs, labor demand, regulation, and access to training.
Employers also may use automation to support workers rather than replace them. Software can process records, prepare drafts, or identify patterns. Employees can then spend more time on judgment, customer service, problem-solving, and oversight.
Workers Face Uneven Risks
Automation is unlikely to affect every occupation at the same pace. Jobs with repeated, predictable tasks may face earlier changes. Work that requires physical flexibility, social judgment, care, or complex decisions may be harder to automate fully.
The report’s headline estimate suggests several issues will require attention:
- Which occupations will experience the largest changes.
- Whether employers will retrain current staff or hire new workers.
- How smaller businesses will pay for new systems and training.
- Whether new jobs will appear in the same places as displaced roles.
Geography may also shape the impact. Regions that depend heavily on a narrow group of occupations could face greater disruption. Areas with strong education systems and varied industries may adjust more quickly.
Training Could Shape the Outcome
The 2035 time frame gives governments, schools, and employers a limited period to prepare. Training programs may need to focus on practical skills that complement automated systems. These include technical oversight, communication, analysis, maintenance, and decision-making.
Access will be as important as course design. Workers with limited time or money may struggle to complete lengthy programs. Short courses, paid workplace training, and recognized credentials could help more people take part.
Companies also face choices about how they introduce automation. Rapid deployment may reduce some costs, but it can weaken morale if workers receive little notice or support. A staged approach can give employees time to learn new duties and help managers identify problems.
Forecast Carries Important Limits
Long-term labor estimates are shaped by assumptions. Economic growth, consumer demand, technical performance, and public policy can change before 2035. The 36 million figure should therefore be read as a warning about possible disruption, not a fixed count of future job losses.
Further detail will be needed to judge the scale of the risk. Useful measures include the number of tasks automated, wages in exposed occupations, retraining results, and the rate at which affected workers find new roles.
The central finding is clear: automation may touch tens of millions of jobs within roughly a decade. The key question is not simply how much technology businesses adopt. It is whether institutions can help workers adapt as duties, skills, and hiring needs change.
