Governments are making a dangerous bet on the AI boom - the Economist names the risk

Governments betting fiscal policy and industrial strategy on the AI boom face an underappreciated fragility: the investment cycle is outrunning the evidence base for productivity returns.

In plain terms: The Economist warns that governments worldwide have committed fiscal and industrial resources to the AI boom at a stage where the macroeconomic productivity gains justifying those commitments remain largely unproven. The bet may be right - but the timing risk is real.

The fact: Sovereign wealth funds, national AI strategies, and state co-investment programs have deployed trillions in direct and indirect exposure to AI infrastructure. This is happening faster than any comparable technology cycle, at a stage where GDP-level productivity data remains ambiguous. The AI power wall, debt-financed capex, and absence of clear macro statistics compound the picture.

Our read: The Economist's concern is structural, not pessimistic about AI's ultimate value. Every major technology cycle has produced real economic gains - eventually. The question is timing: if the productivity inflection is 5 years out rather than 2, the debt maturities, election cycles, and political patience may not hold. For policymakers, the risk isn't that AI fails. It's that the timeline is longer than the instruments financing the bet.

Watch for: G7 AI fiscal commitment revisions as returns lag expectations, IMF or BIS-level assessments of systemic AI exposure, and any major sovereign wealth fund reducing its AI infrastructure allocation. These would be the early warning signals that the policy bet is being repriced.