Can economies automate and employ at the same time?

One of the most important economic debates today is not about artificial intelligence itself. It is about what AI and automation mean for jobs.
Across the world, governments are trying to achieve two objectives at the same time. They want to create employment opportunities for growing populations and young people entering the labour market. At the same time, they want businesses to become more productive, more competitive, and more innovative through automation, robotics, and digital technologies.
The question is straightforward but difficult: can economies create more jobs while increasingly relying on machines and artificial intelligence?
This debate is far from new. More than 200 years ago, workers feared that the machines of the Industrial Revolution would take away their livelihoods. Similar concerns emerged with the rise of assembly lines, computers, and later the internet. In every case, some occupations disappeared while new industries and professions emerged. Over time, technology raised productivity, expanded economic output, and improved living standards.
This historical experience led many economists to argue that technology does not eliminate work; it changes the nature of work.
However, the rise of artificial intelligence has revived the debate because this technology is different from previous waves of automation. Earlier technologies mainly replaced physical labour. AI is increasingly capable of performing cognitive tasks such as analysing data, drafting reports, reviewing contracts, processing information, and supporting decision-making.
The scale of the potential impact is significant. The International Monetary Fund estimates that around 40 per cent of jobs worldwide could be affected by AI, rising to nearly 60 per cent in advanced economies. Goldman Sachs has suggested that up to 300 million jobs globally are exposed to AI-driven automation. Yet exposure does not necessarily mean elimination. The World Economic Forum projects that while technological change may displace around 92 million jobs by 2030, it could also create approximately 170 million new ones, resulting in a net gain of about 78 million jobs.
These figures suggest that the future is unlikely to be defined by mass unemployment. Instead, it will be shaped by how economies manage the transition.
A key point often overlooked in public discussions is that not all automation is the same. Economists increasingly distinguish between two forms of automation.
The first is Labour-Substituting Automation, where technology directly replaces workers. The second is Labour-Augmenting Automation, where technology increases the productivity and value of workers rather than replacing them. This distinction may determine whether automation becomes a threat to employment or a driver of economic opportunity.
Consider a simple example.
A traditional factory may employ 1,000 workers. A modern smart factory producing the same output may require only 300 workers because robots, sensors, and AI systems perform many routine tasks.
If the analysis stops there, the conclusion seems obvious: automation has eliminated 700 jobs.
Yet the reality may be very different.
If that smart factory becomes the centre of a broader industrial ecosystem, it can generate demand for maintenance services, industrial software development, data analytics, logistics operations, supply-chain management, packaging companies, export services, quality-control laboratories, and industrial AI providers.
In this scenario, the factory itself may employ fewer people directly, but it can support thousands of jobs indirectly. The true employment impact lies not only within the factory walls but across the wider economic network built around it.
This brings us to the heart of the debate.
For decades, economic success was often measured by two different indicators: how many jobs an economy creates and how productive those jobs are. Most of the time, the two moved in the same direction. Today, however, artificial intelligence is forcing policymakers to confront a more difficult reality: productivity and employment do not always grow together.
A highly automated factory may produce twice as much output with half the workforce. From an economic perspective, that is a success. Productivity rises, costs fall, exports become more competitive, and profits increase. Yet from a labour-market perspective, fewer people may benefit directly from that growth.
This creates a dilemma that many economies are beginning to face. Is the goal to maximise employment, even if productivity remains modest? Or is the goal to maximise productivity, even if fewer workers are required?
The answer cannot be found at either extreme. An economy that focuses only on job creation may end up supporting low-value activities that struggle to compete globally. An economy that focuses only on productivity may generate impressive economic statistics while leaving growing numbers of people disconnected from the benefits of growth.
The challenge, therefore, is not choosing between jobs and productivity. It is ensuring that productivity growth creates new opportunities rather than simply reducing labour demand.
This is why a growing number of economists and development experts are promoting the concept of Employment-Intensive Value Chains.
The idea is simple but powerful. Rather than measuring jobs only inside a factory or industrial project, policymakers should look at the entire value chain surrounding it. Manufacturing is only one part of the picture. Logistics, maintenance, business services, technical training, software development, marketing, exports, local suppliers, research, and innovation all contribute to employment generation.
Under this model, a factory becomes more than a production facility. It becomes a platform for creating an ecosystem of economic activity and employment.
Ultimately, the debate is not about choosing between automation and jobs. It is about finding the right balance between productivity and employment.
Labour productivity remains one of the most important drivers of long-term economic growth, higher wages, and international competitiveness. At the same time, employment rates, youth unemployment, labour-force participation, and workforce skills remain essential indicators of economic health and social stability.
The economies that succeed in the AI era are unlikely to be those that automate the fastest, nor those that create the largest number of jobs regardless of quality. The real winners will be those that combine technological progress with human development.
History suggests that resisting technology is rarely a successful strategy. Yet history also shows that technology alone does not guarantee prosperity. The challenge for policymakers is to ensure that innovation creates more opportunities than it removes.
That may prove to be one of the defining economic tests of the twenty-first century.