A World Without Work

1. Technological progress threatens widespread unemployment

The threat of 'technological unemployment' is now real.

Accelerating capabilities. Rapid advances in artificial intelligence and robotics are enabling machines to perform an ever-expanding range of tasks previously done by humans. This includes both manual and cognitive work across many industries.

Economic disruption. As machines become more capable, they are likely to displace large numbers of workers, potentially leading to widespread unemployment. This could fundamentally disrupt labor markets and economic structures.

Societal impact. Technological unemployment threatens not just individual livelihoods, but social stability and meaning. As work has been central to human life and society, its loss could have profound psychological and cultural effects.

2. Past anxiety about automation was often misplaced

Throughout history, there has always been enough demand for the work of human beings to avoid the emergence of large pools of permanently displaced people.

Historical precedent. Previous waves of automation, from the Industrial Revolution onward, sparked fears of mass unemployment. However, new jobs and industries consistently emerged to employ displaced workers.

Complementary effects. While machines displaced some workers, they often made others more productive. New technologies also grew the overall economic pie, creating more total demand for labor.

Changing nature of work. Rather than eliminating human work entirely, automation typically changed the types of jobs available. Workers adapted by developing new skills for emerging roles.

3. AI capabilities have advanced rapidly through pragmatic approaches

Despite the initial burst of optimism and enthusiasm, no serious progress was made in AI. When it came to the grand challenges – building a machine with a mind, one that was conscious, or that could think and reason like a human being – the defeat was emphatic.

Shift in approach. Early AI research aimed to replicate human intelligence, with little success. Modern AI has made breakthroughs by using statistical methods and vast datasets to solve specific problems.

Diverse capabilities. Today's AI systems can outperform humans at complex games, recognize speech and images, translate languages, and more. This is despite not replicating human cognitive processes.

Ongoing expansion. AI capabilities continue to grow rapidly across domains like healthcare, finance, transportation, and creative work. The full potential remains unclear but significant.

4. Task encroachment will gradually reduce human employment

As time passes, that work is likely to sit beyond the reach of more and more people. And, as we move through the twenty-first century, the demand for the work of human beings is likely to wither away, gradually.

Incremental displacement. Machines are taking on more tasks over time, slowly reducing the range of activities where humans have an advantage. This is occurring across manual, cognitive, and even social domains.

Uneven impacts. Task encroachment affects different industries, regions, and skill levels at varying rates. Some areas face rapid automation while others change more slowly.

• High-skill knowledge work is increasingly affected
• Many low-skill service jobs remain challenging to automate
• Middle-skill jobs face significant pressure

Long-term trend. While not an overnight shift, the gradual nature of task encroachment makes it a persistent force reshaping employment over decades.

5. Frictional technological unemployment is already emerging

Today is not the first time that automation anxiety has spread, nor did it first appear in the 1930s with Keynes. In fact, ever since modern economic growth began, centuries ago, people have periodically suffered from bouts of intense panic about being replaced by machines.

Skills mismatch. Many workers lack the skills for available jobs, especially as middle-skill roles decline and high-skill jobs grow. Retraining is often challenging.

Identity mismatch. Some workers are unwilling to take available jobs that don't match their self-image or desired lifestyle. This particularly affects displaced manufacturing workers.

Place mismatch. Job opportunities are often concentrated in different geographic areas than where unemployed workers live. Relocation can be difficult.

• Financial constraints limit mobility
• Family and community ties keep people in place
• Housing costs in high-opportunity areas create barriers

6. Structural technological unemployment looms in the long-term

As machines keep becoming increasingly capable, many human beings will eventually be driven out of work.

Weakening complementary effects. While automation has historically complemented human labor in many ways, these positive effects may diminish as machines become more broadly capable.

Shrinking human advantage. The range of tasks where humans clearly outperform machines is likely to keep narrowing, leaving fewer roles where human workers are essential.

Economic shifts. As the relative productivity and cost-effectiveness of machines improve, economic incentives will increasingly favor automation over human employment in many areas.

• Capital owners benefit more than workers
• Labor's share of economic output may continue to decline
• New job creation may not keep pace with displacement

7. Rising inequality foreshadows technological unemployment

Today, we talk about 'horsepower', harking back to a time when the pulling power of a draft horse was a measure that mattered, future generations may come to use the term 'manpower' as a similar kind of throwback, a relic of a time when human beings considered themselves so economically important that they crowned themselves as a unit of measurement.

Income concentration. Wealth and income are increasingly concentrated among top earners and capital owners, while wages stagnate for many workers.

Labor market polarization. The job market is splitting between high-skill, high-wage jobs and low-skill, low-wage jobs, with a hollowing out of the middle class.

Capital vs. labor. The share of economic output going to workers (wages) is declining relative to the share going to capital owners (profits, rents, etc.).

• Top 1% income share has roughly doubled in many countries since 1980
• Labor's share of GDP has fallen significantly in most developed economies
• Wealth inequality is even more extreme than income inequality

8. Education alone cannot solve technological unemployment

More education was needed.

Limits of retraining. While education and retraining are valuable, they face limits in preparing all workers for an automated economy. Some lack the capability or resources for high-skill roles.

Changing skill demands. The skills needed in the economy are evolving rapidly, making it difficult for education systems to keep pace and provide relevant training.

Insufficient job creation. Even with improved education, there may simply not be enough jobs for all workers as automation progresses. Education can't solve a fundamental lack of demand for human labor.

• Automation is affecting even high-skill, well-educated workers
• The pace of technological change outstrips many workers' ability to adapt
• Social and economic barriers limit access to effective retraining for many

9. A "Big State" will be needed to address unemployment and inequality

If we are to find a way to narrow the inequalities by a less cataclysmic route than in the past, it is clear that tinkering and tweaking, as the state has tried before, will not be enough. The only way to deal with the looming disparities is to attack them aggressively and directly.

Expanded redistribution. Governments will likely need to significantly increase taxes on high earners and capital owners to fund expanded social programs and income supports.

Universal basic income. Some form of guaranteed income may be necessary to ensure basic living standards as employment declines. This could be unconditional or tied to social contributions.

Public employment. The state may need to become an employer of last resort, creating public service jobs to employ those displaced by automation.

• Higher taxes on capital, high incomes, and wealth
• Expanded social safety net and public services
• Potential job guarantee or work-sharing programs

10. Big Tech's growing power requires new regulatory approaches

In the twenty-first century, we will have to build a new age of security, one that no longer relies on paid work for its foundations. And we have to begin this task today.

Economic concentration. A small number of technology companies are gaining enormous market power and financial resources. This raises concerns about competition and innovation.

Social and political influence. Big Tech firms increasingly shape public discourse, access to information, and even democratic processes. This creates new challenges for governance and accountability.

Data and privacy issues. The vast data collection and analysis capabilities of tech giants raise significant privacy concerns and questions about data rights.

• New antitrust approaches may be needed for the digital economy
• Regulation of social media and content platforms is evolving
• Data protection and algorithmic accountability are key policy areas

11. Finding meaning beyond work will be a key challenge

We can speculate about what those required activities will be. Some communities, populated by people like Keynes and Russell, might be satisfied if those without work spend their time in pursuit of artistic and cultural ends: reading, writing, composing beautiful music, thinking deep thoughts.

Redefining purpose. As work becomes less central to many people's lives, individuals and societies will need to find new sources of meaning, identity, and social connection.

Cultural shift. Moving away from the work-centric culture that has dominated since the Industrial Revolution will require significant psychological and social adaptation.

New forms of contribution. Non-economic activities like volunteering, caregiving, civic engagement, and creative pursuits may gain greater social recognition and support.

• Education may need to focus more on life skills and personal development
• Community organizations could play a larger role in providing structure and purpose
• Universal basic income might be tied to social contributions rather than traditional work

Last updated: November 4, 2024