Taiwan’s government has set aside NT$235.7 billion, or $7.4 billion, in its 2027 budget to pay AI dividends to citizens. This amounts to 10,000 NT$, or 314 dollars, per person.
What is behind such generosity?
Rising global demand for AI, high-performance computing, and cloud services is boosting orders, exports, and investment, pushing Taiwan’s economy into double-digit growth for the third quarter in a row, with first-half growth reaching 14.15%. Hence, the government decided it was only fair to share some of the gains with the public.
The U.S. economy is also getting a boost from AI investment, but so far, investors are capturing most of the gains, while ordinary people are bearing the costs of data centres and higher electricity prices. As for jobs lost to AI, they should, in theory, be offset by new jobs in the sector, although July’s numbers showed the replacement is still not complete.
Europe should have the hardest time. The region is far behind the U.S. and China in AI investment due to AI regulation, expensive electricity and complex rules for building data centres, so it is not only missing out on AI investment but could also lose jobs without enough new ones to offset them.
The reassuring part is that if AI turns out to be another bubble, Europe would also feel the impact, though probably less severely than the Dow Jones, Nasdaq or S&P 500 indices in the U.S.
As for how likely a bubble is, there is one major difference from the dot-com crisis 25 years ago: companies are actually making money from AI. Microsoft, for example, has surpassed $100 billion in annual Azure revenue; Oracle is nearly doubling its cloud business; Amazon is seeing AWS growth accelerate; and Celestica and AMD are confirming strong demand for AI infrastructure.
However, it’s worth noting that the WSJ found nine major tech companies have accumulated around $3 trillion in off-balance-sheet commitments, locking in huge future expenses on the assumption that AI demand will keep growing. If, in turn, demand falls short, tech giants could be left with excess data centres, expensive contracts and the need to borrow more.
A mistake in forecasting demand could turn the AI investment supercycle into serious pressure for financial markets, including Europe, though the impact could be less severe due to its lower exposure.