Latest / Can't Find Job? AI Is Quietly Replacing Millions of Workers / Scenarios 2026–2030: Jobs, AI Diffusion, and Productivity Paths

Transcript

Scenarios 2026–2030: Jobs, AI Diffusion, and Productivity Paths · Can't Find Job? AI Is Quietly Replacing Millions of Workers

• 0:00
  SPEAKER_00: Scenarios 2026 to 2030, jobs, artificial intelligence diffusion, and productivity paths introduction.
• 0:08
  SPEAKER_00: The global labor market is entering a period of rapid technological change, but uneven economic adjustment.
• 0:15
  SPEAKER_00: Artificial intelligence is spreading quickly through business software, customer service, finance, marketing, software development, healthcare, logistics, and public services.
• 0:25
  SPEAKER_00: Yet the labor market evidence does not show one simple outcome.
• 0:29
  SPEAKER_00: So far, artificial intelligence has improved productivity in many tasks, but it has not caused broad global unemployment.
• 0:36
  SPEAKER_00: At the same time, early career workers in some exposed occupations are facing weaker hiring, real wage growth is slowing, and the share of national income going to labor has fallen in some economies.
• 0:47
  SPEAKER_00: This article develops three evidence-based scenarios for 2026 to 2030.
• 0:52
  SPEAKER_00: High augmentation.
• 0:54
  SPEAKER_00: Artificial intelligence mainly helps workers produce more and creates new demand.
• 0:59
  SPEAKER_00: Mixed.
• 0:59
  SPEAKER_00: Adoption spreads but productivity gains are uneven and labor market adjustment is slow.
• 1:04
  SPEAKER_00: Disruption.
• 1:05
  SPEAKER_00: Artificial intelligence replaces large numbers of routine tasks faster than workers, firms, and governments can create new opportunities.
• 1:12
  SPEAKER_00: The scenarios are calibrated against labor market, wage, productivity, adoption, investment, and regulatory data available through July 2026.
• 1:22
  SPEAKER_00: The latest United States Employment Report covers July 2026, while the latest job openings and labor turnover survey covers June 2026 because the July survey is scheduled for release on September 1, 2026.
• 1:36
  SPEAKER_00: Executive Summary.
• 1:37
  SPEAKER_00: The mixed scenario is the most likely central path.
• 1:40
  SPEAKER_00: Artificial intelligence adoption will continue to rise, but many firms will use it for narrow tasks rather than redesigning entire organizations.
• 1:48
  SPEAKER_00: Productivity will improve, but the gains will be concentrated in firms with strong data, management, computing access, and skilled workers.
• 1:56
  SPEAKER_00: The high augmentation scenario is achievable if firms invest in training, redesign jobs around human-machine cooperation, and share productivity gains through higher wages, shorter work hours, or expanded output.
• 2:08
  SPEAKER_00: The disruption scenario is a serious downside risk, especially for entry-level office work, customer support, routine finance, administrative services, basic software work, and some professional services.
• 2:19
  SPEAKER_00: It becomes more likely if the cost of artificial intelligence falls rapidly, artificial intelligence agents become reliable, and firms use the technology mainly to reduce payroll rather than expand production.
• 2:30
  SPEAKER_00: For the full table, please open this article on can't findjob.com.
• 2:35
  SPEAKER_00: These job ranges are scenario estimates, not official forecasts.
• 2:39
  SPEAKER_00: They measure the possible effect of artificial intelligence and related organizational change, not total employment from population growth, healthcare demand, the green transition, or other forces.
• 2:51
  SPEAKER_00: What the evidence shows through July 2026, labor markets are cooling, but they have not collapsed.
• 2:57
  SPEAKER_00: The International Labor Organization expects global unemployment to remain near 4.9% in 2026.
• 3:04
  SPEAKER_00: However, the broader global jobs gap, which includes people who want paid work but cannot find it, is projected at 408 million people.
• 3:13
  SPEAKER_00: About 2.1 billion workers remain in informal employment, showing that the headline unemployment rate does not capture the full weakness of global labor markets.
• 3:22
  SPEAKER_00: The United States provides a useful high-frequency benchmark.
• 3:26
  SPEAKER_00: The number of unemployed people per job opening was about 0.8 to 0.9 in 2019, rose to 5.0 in April 2020, fell to 0.5 during 2022, and moved back to about 1.0 in May 2026.
• 3:41
  SPEAKER_00: The ratio has remained between 0.9 and 1.1 since April 2024, which suggests a labor market that has normalized from the extreme shortage of workers in 2021 and 2022, but is not yet deeply weak.
• 3:56
  SPEAKER_00: In June 2026, the United States had 7.4 million job openings, a job opening rate of 4.4%, 5.3 million hires, and 1.8 million layoffs and discharges.
• 4:09
  SPEAKER_00: In July 2026, payroll employment fell by 23,000, the unemployment rate was 4.1%, and average private sector hourly earnings were up 3.2% from a year earlier.
• 4:21
  SPEAKER_00: Healthcare continued to add jobs while retail trade, local government education, finance, and leisure and hospitality weakened.
• 4:29
  SPEAKER_00: The European Union also shows easing demand.
• 4:32
  SPEAKER_00: Its job vacancy rate was 2.1% in the first quarter of 2026 compared with 2.2% a year earlier.
• 4:39
  SPEAKER_00: The Euro area rate was 2.3%.
• 4:42
  SPEAKER_00: Artificial intelligence adoption has moved from experimentation to diffusion.
• 4:48
  SPEAKER_00: Survey evidence shows a large change since the late 2010s.
• 4:52
  SPEAKER_00: McKinsey surveys found that artificial intelligence adoption remained near one half of organizations for several years, then rose to 72% in early 2024.
• 5:03
  SPEAKER_00: In the latest survey cited by the company, 78% of respondents said their organizations used artificial intelligence in at least one business function.
• 5:12
  SPEAKER_00: The definitions and samples changed over time, so these figures should be treated as adoption signals rather than perfectly comparable national statistics.
• 5:21
  SPEAKER_00: The 2026 Stanford Artificial Intelligence Index reports that 88% of surveyed organizations used artificial intelligence in 2025.
• 5:30
  SPEAKER_00: While generative artificial intelligence was used in at least one business function by 70% of organizations.
• 5:37
  SPEAKER_00: However, artificial intelligence agent use remained in the single digits across nearly all business functions.
• 5:44
  SPEAKER_00: This is important.
• 5:45
  SPEAKER_00: Many firms have adopted artificial intelligence tools, but far fewer have allowed them to complete complex workflows with limited human supervision.
• 5:54
  SPEAKER_00: Adoption is also uneven across countries.
• 5:57
  SPEAKER_00: An international monetary fund study using observed artificial intelligence usage from January 2025 through February 2026 found that developing economies often concentrated artificial intelligence-related value in a small professional group.
• 6:13
  SPEAKER_00: The study estimated that the time currently saved by artificial intelligence had a labor cost equivalent of about US$2.7 trillion annually, or approximately 3.4% of global gross domestic product.
• 6:28
  SPEAKER_00: This is not the same as realized economic output, but it shows that current use is already economically meaningful.
• 6:34
  SPEAKER_00: Capital investment is accelerating.
• 6:37
  SPEAKER_00: Artificial intelligence investment has expanded much faster than most earlier digital technologies.
• 6:43
  SPEAKER_00: Stanford reports that global corporate artificial intelligence investment more than doubled in 2025.
• 6:50
  SPEAKER_00: Private investment increased by 127.5%, while generative artificial intelligence investment grew by more than 200%.
• 6:58
  SPEAKER_00: Google reported more than US$150 billion in annual capital spending in 2025, reflecting the growing cost of data centers, advanced chips, electricity, and networking infrastructure.
• 7:12
  SPEAKER_00: The main question is whether this spending will spread beyond a small number of large technology companies.
• 7:18
  SPEAKER_00: A high augmentation path requires investment in business process redesign, worker training, data quality, cybersecurity, electricity and telecommunications, small and medium-sized businesses, public sector systems, and local language tools for developing economies.
• 7:34
  SPEAKER_00: If investment remains concentrated in computing infrastructure without organizational change, productivity gains will remain narrow.
• 7:42
  SPEAKER_00: Productivity is improving, but labor income is under pressure.
• 7:46
  SPEAKER_00: United States non-farm business productivity grew 2.2% from the second quarter of 2025 to the second quarter of 2026.
• 7:55
  SPEAKER_00: Since the fourth quarter of 2019, productivity has grown at an annualized rate of 2.1%, above the 1.5% rate of the previous business cycle and equal to the long-term average.
• 8:08
  SPEAKER_00: However, the distribution of those gains is a concern.
• 8:11
  SPEAKER_00: The labor share of income in the United States fell to 52.9% in the second quarter of 2026, the lowest point in the series that begins in 1947.
• 8:22
  SPEAKER_00: Real hourly compensation fell by 3.1% at an annual rate during that quarter.
• 8:27
  SPEAKER_00: Nominal wage growth remains positive.
• 8:29
  SPEAKER_00: The Atlanta Federal Reserve's wage tracker showed median wage growth of 3.6% in June 2026, with wage growth of 4.1% for people changing jobs and 3.4% for people staying in the same job.
• 8:44
  SPEAKER_00: Across the Organization for Economic Cooperation and Development, real wage growth averaged 2.2% in the first quarter of 2026, down from 2.7% a year earlier.
• 8:56
  SPEAKER_00: Real wages remained below early 2021 levels in about one-third of member countries.
• 9:02
  SPEAKER_00: Regulation will influence the speed and form of adoption.
• 9:06
  SPEAKER_00: The European Union Artificial Intelligence Act is a major regulatory timetable to watch.
• 9:12
  SPEAKER_00: Prohibited practices and artificial intelligence literacy requirements began applying on February 2, 2025.
• 9:20
  SPEAKER_00: Governance rules and obligations for general-purpose artificial intelligence models began applying on August 2, 2025.
• 9:28
  SPEAKER_00: The general application date is August 2, 2026.
• 9:32
  SPEAKER_00: Following the 2026 digital omnibus changes, some high-risk rules have longer transition periods.
• 9:39
  SPEAKER_00: High-risk systems embedded in regulated products have an extended transition period until August 2, 2028, while certain high-risk use cases in sensitive areas have an extended period until December 2, 2027.
• 9:53
  SPEAKER_00: Regulation will not stop diffusion, but it can affect the cost of compliance, the speed of deployment in healthcare and finance, public sector procurement, worker monitoring, automated hiring and dismissal, liability for errors, and demand for artificial intelligence auditors and compliance specialists.
• 10:10
  SPEAKER_00: Scenario Framework.
• 10:12
  SPEAKER_00: The scenarios use four main variables.
• 10:14
  SPEAKER_00: Diffusion, how widely artificial intelligence spreads beyond large technology companies.
• 10:19
  SPEAKER_00: Task substitution, whether tools assist workers or replace their tasks.
• 10:24
  SPEAKER_00: Demand expansion, whether lower costs create enough new demand to generate more jobs.
• 10:29
  SPEAKER_00: Adjustment capacity, whether workers, firms, schools, and governments can move people into growing occupations.
• 10:35
  SPEAKER_00: The World Economic Forum's 2025 Employer Survey provides a useful middle reference point.
• 10:41
  SPEAKER_00: It estimated that 170 million jobs could be created and 92 million displaced by 2030 from several large economic trends, producing a net increase of 78 million jobs.
• 10:53
  SPEAKER_00: It also found that about 40% of required job skills may change, and that 63% of employers viewed skill gaps as a major barrier to business transformation.
• 11:04
  SPEAKER_00: The three scenarios below build around that middle reference point.
• 11:08
  SPEAKER_00: Scenario 1, high augmentation, core idea.
• 11:11
  SPEAKER_00: In this scenario, artificial intelligence becomes a general productivity tool rather than a mass replacement system.
• 11:18
  SPEAKER_00: Workers use artificial intelligence to prepare information, draft documents, analyze data, translate material, monitor equipment, design products, and support decisions.
• 11:28
  SPEAKER_00: Humans remain responsible for judgment, relationships, safety, accountability, and complex problem solving.
• 11:34
  SPEAKER_00: Firms redesign jobs instead of simply cutting headcount.
• 11:38
  SPEAKER_00: Productivity gains lower prices, expand demand, and create new products and services.
• 11:44
  SPEAKER_00: Governments support training, competition, infrastructure, and worker mobility.
• 11:49
  SPEAKER_00: Labor market outcome employment remains strong because productivity gains lead to more output and new demand.
• 11:56
  SPEAKER_00: The largest employment gains occur in healthcare, construction, energy, education, technical services, software, cybersecurity, and industries that use artificial intelligence to expand capacity.
• 12:08
  SPEAKER_00: The main risk is not unemployment, but unequal access.
• 12:12
  SPEAKER_00: Workers and firms with strong education, data, infrastructure, and management improve quickly.
• 12:18
  SPEAKER_00: Others fall behind.
• 12:19
  SPEAKER_00: Sectoral Employment Projection, estimated change in employment by 2030 compared with 2025 employment levels.
• 12:26
  SPEAKER_00: For the full table, please open this article on can'tfindjob.com.
• 12:31
  SPEAKER_00: These estimates reflect artificial intelligence along with aging, infrastructure demand, energy investment, and the continued shift from agriculture towards services.
• 12:41
  SPEAKER_00: The International Labor Organization estimates that about 1 billion people work in agriculture, equal to approximately 28% of global employment, so changes in agricultural productivity can have large effects even when the share of agricultural employment continues to fall.
• 12:57
  SPEAKER_00: Skill demand.
• 12:59
  SPEAKER_00: The fastest growing skills are likely to include artificial intelligence literacy, data interpretation, workflow design, cybersecurity, technical maintenance, product design, communication and teamwork, teaching and coaching, healthcare and personal care, judgment and quality control, energy and computing infrastructure.
• 13:18
  SPEAKER_00: The most valuable workers will combine technical ability with knowledge of a real industry.
• 13:22
  SPEAKER_00: A health worker who understands artificial intelligence tools may be more valuable than a general technology worker who does not understand clinical work.
• 13:31
  SPEAKER_00: Policy needs.
• 13:32
  SPEAKER_00: Governments should fund basic digital and artificial intelligence training for adults, expand apprenticeships in construction, energy, healthcare, and technology, help small businesses adopt safe productivity tools, invest in broadband, electricity, data systems, and local language technology, require employers to consult workers before major automated changes.
• 13:53
  SPEAKER_00: Support wage growth through stronger worker bargaining power and productivity sharing agreements.
• 13:58
  SPEAKER_00: Make benefits portable for workers who change jobs or work across several platforms.
• 14:03
  SPEAKER_00: The key policy goal is to make artificial intelligence widely usable, not merely widely available.
• 14:10
  SPEAKER_00: Scenario 2.
• 14:11
  SPEAKER_00: Mixed adoption.
• 14:12
  SPEAKER_00: Core idea.
• 14:13
  SPEAKER_00: This is the most likely central scenario.
• 14:16
  SPEAKER_00: Artificial intelligence spreads rapidly, but adoption remains uneven.
• 14:20
  SPEAKER_00: Large firms automate more tasks while smaller firms use off-the-shelf tools.
• 14:24
  SPEAKER_00: Some managers redesign jobs around artificial intelligence.
• 14:28
  SPEAKER_00: Productivity rises, but the gains are not large enough to create a broad employment boom.
• 14:34
  SPEAKER_00: Worker transitions are slow because training systems, schools, and local labor markets do not adjust quickly.
• 14:40
  SPEAKER_00: Labor market outcome.
• 14:42
  SPEAKER_00: The labor market becomes more divided.
• 14:44
  SPEAKER_00: High-skill workers who can supervise or improve artificial intelligence systems gain bargaining power.
• 14:50
  SPEAKER_00: Middle skill workers face pressure when routine tasks are automated.
• 14:54
  SPEAKER_00: Entry-level workers lose some traditional pathways into professional occupations.
• 14:58
  SPEAKER_00: Healthcare, construction, education, and skilled trades remain relatively strong.
• 15:04
  SPEAKER_00: Workers in declining regions have difficulty moving into growing industries.
• 15:08
  SPEAKER_00: The Organization for Economic Cooperation and Development estimates that the share of workers exposed to generative artificial intelligence ranges from about 16% in some regions to more than 70% in others.
• 15:21
  SPEAKER_00: This wide range means that the same technology can produce very different local outcomes.
• 15:26
  SPEAKER_00: Sectoral Employment Projection.
• 15:28
  SPEAKER_00: For the full table, please open this article on can'tfindjob.com.
• 15:32
  SPEAKER_00: Skill Demand.
• 15:34
  SPEAKER_00: Demand rises for artificial intelligence-supported office work, data quality and data management, compliance and risk management, customer relationship skills, skilled trades, healthcare, education and training, cybersecurity, project management, artificial intelligence testing, and quality assurance.
• 15:51
  SPEAKER_00: Demand falls for jobs based mainly on repetitive document preparation, basic research, simple scheduling, routine customer support, data entry, standardized bookkeeping, basic translation, simple software coding.
• 16:06
  SPEAKER_00: A key skill is adaptability.
• 16:08
  SPEAKER_00: Workers will need to learn new tools several times during their careers, not once.
• 16:12
  SPEAKER_00: Policy needs.
• 16:14
  SPEAKER_00: Governments should prepare for frequent job changes rather than waiting for mass unemployment.
• 16:18
  SPEAKER_00: Useful policies include training accounts that workers can use throughout their careers, short courses tied directly to local vacancies, wage insurance for workers who accept lower-paid jobs after displacement, better job matching services, housing and transportation support for workers who move to stronger regions, public funding for community colleges and technical schools, clear rules for automated hiring, worker monitoring and performance scoring, stronger measurement of artificial intelligence exposure by occupation and region.
• 16:48
  SPEAKER_00: The most important policy shift is from passive unemployment support to early transition support.
• 16:54
  SPEAKER_00: Scenario 3.
• 16:55
  SPEAKER_00: Disruption.
• 16:56
  SPEAKER_00: Core idea.
• 16:57
  SPEAKER_00: In this scenario, artificial intelligence becomes reliable and inexpensive enough to replace large numbers of routine cognitive tasks.
• 17:05
  SPEAKER_00: Artificial intelligence agents can handle customer service, scheduling, research, document production, claims processing, basic legal work, routine financial analysis, software testing, and some management tasks.
• 17:19
  SPEAKER_00: Firms adopt the technology faster than education systems and labor markets can adjust.
• 17:25
  SPEAKER_00: Productivity rises sharply in leading firms, but demand does not expand enough to absorb displaced workers.
• 17:32
  SPEAKER_00: Labor market outcome.
• 17:33
  SPEAKER_00: A disruption scenario would probably appear first in hiring data, rather than total employment data.
• 17:39
  SPEAKER_00: Early warning signs would include falling entry-level hiring, longer job searches, declining vacancies in exposed occupations, rising long-term unemployment, lower wage growth for young workers, more temporary and contract work, a widening gap between company productivity and worker pay.
• 17:58
  SPEAKER_00: The 2026 Stanford Artificial Intelligence Index already reports that employment for software developers aged 22 to 25 had fallen nearly 20% from 2024.
• 18:09
  SPEAKER_00: It also reports that one-third of surveyed organizations expected artificial intelligence to reduce their workforce during the following year, although broad job losses had not yet appeared in overall employment data.
• 18:21
  SPEAKER_00: Sectoral Employment Projection, for the full table, please open this article on can'tfindjob.com.
• 18:27
  SPEAKER_00: The largest losses would probably occur in routine office and service work.
• 18:32
  SPEAKER_00: Some professional occupations would also be affected, especially at the entry level.
• 18:36
  SPEAKER_00: Technical jobs connected to computing infrastructure, security, model testing, and high-value product development would continue to grow.
• 18:45
  SPEAKER_00: Skill demand the labor market would place a premium on complex human relationships, physical work and changing environments, safety and accountability, advanced technical supervision, artificial intelligence system management, entrepreneurship, skilled construction and maintenance, health care and personal care, teaching and community support, conflict resolution.
• 19:05
  SPEAKER_00: Workers would also need transition skills, the ability to move quickly from a declining occupation into a growing one.
• 19:12
  SPEAKER_00: Policy needs.
• 19:13
  SPEAKER_00: The disruption scenario requires stronger action, more generous unemployment and wage replacement support, fast retraining for workers in exposed occupations, public employment and infrastructure programs in weak regions, portable health care, retirement and paid leave benefits, stronger rules against discriminatory automated hiring and dismissal, worker representation and decisions about automated systems, possible reductions in working hours if productivity rises faster than labor demand.
• 19:41
  SPEAKER_00: Serious discussion of how to distribute income from highly automated production.
• 19:46
  SPEAKER_00: The main danger is not only unemployment, it is a loss of economic security and bargaining power for people whose work remains available but becomes poorly paid.
• 19:56
  SPEAKER_00: Sectoral patterns common to all three scenarios.
• 20:00
  SPEAKER_00: Some trends are likely in every scenario.
• 20:03
  SPEAKER_00: Healthcare and care work grow.
• 20:05
  SPEAKER_00: Aging populations, rising incomes, and unmet care needs will support employment in nursing, home care, mental health, disability services, and health administration.
• 20:16
  SPEAKER_00: Artificial intelligence may reduce paperwork, but it is unlikely to remove the need for trust, physical assistance, empathy, and human responsibility.
• 20:25
  SPEAKER_00: Construction and energy remain important.
• 20:28
  SPEAKER_00: Electricity generation, data centers, housing, transport, climate adaptation, and industrial investment will support construction and technical maintenance.
• 20:37
  SPEAKER_00: These sectors may adopt artificial intelligence, but much of the work remains physical and location specific.
• 20:44
  SPEAKER_00: Routine office work faces the most pressure.
• 20:46
  SPEAKER_00: Administrative, finance, customer service, and basic research tasks are highly exposed because their outputs are often digital, measurable, and repetitive.
• 20:56
  SPEAKER_00: The Organization for Economic Cooperation and Development estimates that occupations at high risk of automation account for about 27% of employment in its member countries when artificial intelligence is included with other automation technologies.
• 21:10
  SPEAKER_00: Agriculture changes through productivity more than job creation.
• 21:14
  SPEAKER_00: Artificial intelligence, robotics, sensors, and improved forecasting may raise output while reducing labor needs in some regions.
• 21:22
  SPEAKER_00: In other regions, population growth and low farm productivity will continue to create agricultural employment.
• 21:28
  SPEAKER_00: The global result will be uneven.
• 21:31
  SPEAKER_00: Software work divides into two groups.
• 21:33
  SPEAKER_00: Demand will grow for advanced system design, security, infrastructure, data engineering, and artificial intelligence evaluation.
• 21:41
  SPEAKER_00: Demand may weaken for basic coding, routine testing, and simple software maintenance.
• 21:46
  SPEAKER_00: Leading indicators to monitor.
• 21:48
  SPEAKER_00: For the full table, please open this article on can'tfindjob.com.
• 21:53
  SPEAKER_00: The most important measure is not the number of organizations that say they use artificial intelligence, it is the share of work completed by artificial.
• 22:00
  SPEAKER_00: Artificial intelligence, the share completed jointly by people and machines, and the number of workers moved into higher value tasks.
• 22:07
  SPEAKER_00: Decision points for governments.
• 22:09
  SPEAKER_00: Decision Point 1, 2026, 2027.
• 22:12
  SPEAKER_00: Governments should build a national and regional artificial intelligence labor market dashboard.
• 22:17
  SPEAKER_00: It should track job postings by skill, hiring by age and education, vacancies by occupation, wage growth, layoffs, training completion, internal worker movement, artificial intelligence adoption by firm size, the first priority is measurement.
• 22:32
  SPEAKER_00: Policymakers cannot manage a transition they cannot see.
• 22:36
  SPEAKER_00: Decision point two When entry-level hiring falls, if hiring for young workers in exposed occupations falls by 10% or more for several quarters, governments should activate targeted training, wage insurance, apprenticeships, and public employment support.
• 22:51
  SPEAKER_00: This should happen before long-term unemployment becomes widespread.
• 22:55
  SPEAKER_00: Decision point three, when productivity rises faster than pay, if productivity growth exceeds real wage growth by more than two percentage points for several years, governments should review competition policy, worker bargaining power, profit sharing systems, minimum wage rules, portable benefits, tax treatment of labor and capital, and working time policies.
• 23:16
  SPEAKER_00: Decision point 4, 2028 to 2030.
• 23:19
  SPEAKER_00: Governments should decide whether short-term training programs are enough or whether permanent changes are needed.
• 23:25
  SPEAKER_00: If artificial intelligence continues to reduce the demand for routine work, countries may need broader income protection, shorter working hours, and stronger public services.
• 23:35
  SPEAKER_00: Decision points for firms, map tasks not job titles.
• 23:39
  SPEAKER_00: A job is usually a bundle of tasks.
• 23:41
  SPEAKER_00: Firms should identify which tasks artificial intelligence can assist, automate, improve through better information, leave unchanged, make more valuable because human judgment is needed.
• 23:52
  SPEAKER_00: This prevents the common mistake of treating an entire occupation as either safe or doomed.
• 23:57
  SPEAKER_00: Set an augmentation first policy.
• 23:59
  SPEAKER_00: Before reducing headcount, firms should test whether productivity gains can be used to increase output, improve quality, reduce waiting times, expand into new markets, shorten work hours, move workers into higher value tasks.
• 24:13
  SPEAKER_00: This is more likely to create long-term value than immediate payroll reduction.
• 24:17
  SPEAKER_00: Measure worker outcomes.
• 24:18
  SPEAKER_00: Every major artificial intelligence project should report output per worker, error rates, customer satisfaction, employee workload, training completion, internal transfers, wage changes, number of jobs eliminated, number of jobs created.
• 24:33
  SPEAKER_00: Protect entry-level pathways.
• 24:35
  SPEAKER_00: Firms that eliminate junior jobs may reduce their future supply of experienced workers.
• 24:39
  SPEAKER_00: They should create apprenticeships, supervised artificial intelligence work, and structured learning roles.
• 24:45
  SPEAKER_00: Prepare for regulation.
• 24:46
  SPEAKER_00: Firms should not wait for every legal rule to become final.
• 24:49
  SPEAKER_00: They should already document what data their systems use, how automated decisions are checked, when a human must review an outcome, how workers can challenge decisions, how errors are reported, who is responsible for harm.
• 25:02
  SPEAKER_00: Actionable advice for workers.
• 25:04
  SPEAKER_00: Workers can prepare by building a combination of industry knowledge, artificial intelligence literacy, communication skills, data and quality control skills, problem-solving ability, a record of measurable results.
• 25:16
  SPEAKER_00: The safest career strategy is not simply to learn one software tool.
• 25:21
  SPEAKER_00: It is to become the person who understands a real customer, patient, machine, process, or market, and can use new tools to improve it.
• 25:30
  SPEAKER_00: Workers should also watch local vacancies, not only national headlines.
• 25:35
  SPEAKER_00: Construction, healthcare, education, energy, cybersecurity, and technical maintenance may offer stronger opportunities than routine office work in many regions.
• 25:45
  SPEAKER_00: Conclusion.
• 25:46
  SPEAKER_00: The 2026 to 2030 labor market will not be determined by artificial intelligence alone.
• 25:52
  SPEAKER_00: It will be shaped by how quickly firms adopt it, whether productivity creates new demand, how workers share the gains, and how fast institutions adjust.
• 26:01
  SPEAKER_00: The evidence through July 2026 supports a cautious conclusion.
• 26:06
  SPEAKER_00: Artificial intelligence adoption is accelerating.
• 26:08
  SPEAKER_00: Productivity growth is improving.
• 26:10
  SPEAKER_00: Labor markets remain resilient but are softening.
• 26:14
  SPEAKER_00: Real wage gains are slowing.
• 26:15
  SPEAKER_00: Capital investment is surging.
• 26:17
  SPEAKER_00: Entry-level workers in exposed occupations are showing early signs of pressure.
• 26:22
  SPEAKER_00: Regulation is becoming more important.
• 26:24
  SPEAKER_00: Large-scale job destruction has not yet appeared in overall employment data.
• 26:28
  SPEAKER_00: The most likely future is mixed, but it is not fixed.
• 26:32
  SPEAKER_00: High augmentation can be encouraged through training, worker participation, public infrastructure, competition, and better management.
• 26:40
  SPEAKER_00: Disruption becomes more likely when firms automate faster than economies can create new work.
• 26:45
  SPEAKER_00: The central decision for governments and firms is therefore simple.
• 26:49
  SPEAKER_00: Will artificial intelligence be used mainly to replace people or to help more people produce valuable work?
• 26:55
  SPEAKER_00: All links to sources are available in the text version of this article.
• 26:59
  SPEAKER_00: You can find the full article at can'tfindjob.comslash blog.
• 27:04
  SPEAKER_00: Thanks for listening.
• 27:05
  SPEAKER_00: If today's episode hit close to home, stop scrolling job boards that weren't built for this new reality.
• 27:10
  SPEAKER_00: Check out Claw Earn on AIAagentStore.ai, the first jobs marketplace designed for both humans and AI agents, so you can start earning no matter which side of the AI revolution you're on.