World Bank: AI could add 12 percent to Poland's GDP by 2035

A World Bank report, covered by Forsal.pl, measures two sides of one shift. Widespread adoption of artificial intelligence is expected to raise Poland's GDP by 12 percent by 2035. The structural cost falls very unevenly across industries.

Financial services and insurance are the most exposed. Employment there could fall by nearly 24 percent. The World Bank also expects fewer jobs for programmers and in IT services. Construction sits on the other side of the ledger: the authors expect the number of people in work to rise by nearly 19 percent.

The net figure comes down to one variable: how freely workers can move. Assume high mobility, and the bank puts the fall in employment at 3,000 jobs. Assume low mobility, and the same mechanism leaves 341,000 fewer jobs. That is a gap of two orders of magnitude, and it is the most important number in the report.

The third thread is the budget. The AI shift will cut revenue from personal income tax and from social insurance contributions. In 2026, the bank estimates, that means 0.2 percent and 0.6 percent of GDP respectively. Corporate tax works the other way, adding 0.1 percent of GDP, and VAT adds 0.3 percent.

The Polish report fits a wider picture the OECD drew in its September review. The organisation raised its short-term growth forecasts, but noted that in G20 economies momentum will keep coming from spending and production tied to artificial intelligence, and that this will "often outweigh relatively weak investment in other areas of the corporate sector".

That asymmetry is the interesting part. AI is one of the few engines of private investment that has not slowed, and it is also what is pressing on the labour market and on public finances. For economic policy, counting GDP growth alone says little about what happens to taxes, or to the people in jobs a model can do more cheaply.

The World Bank report goes beyond the aggregate result. Its authors look at the structure of employment and find that workers in financial and insurance services will feel the change most sharply, with employment in that industry possibly falling by nearly 24 percent. Construction sits on the other side, gaining close to 19 percent more jobs, largely through investment in computing and energy infrastructure.

The size of the net effect depends on worker mobility. With high mobility and a smooth flow between sectors, employment barely changes, at about 3,000 jobs. With low mobility, the same scenario means a loss on the order of 341,000 jobs. The policy conclusion is blunt: what decides the outcome is not the technology but the system for retraining people and supporting them through a change of occupation.

The fiscal picture is just as concrete. In 2026, personal income tax revenue would fall by 0.2 percent of GDP and social insurance contributions by 0.6 percent of GDP. In the same scenario, corporate income tax would rise by 0.1 percent of GDP and VAT by 0.3 percent, because some of the workers who lose their jobs would move into other, more consumption-driven work, and the firms deploying the models would book the profits.