Switzerland — The global artificial intelligence investment boom is creating fresh challenges for financial policymakers as spending on AI infrastructure reaches a scale large enough to affect economic conditions, the head of the Bank for International Settlements said.
Pablo Hernández de Cos said AI’s rapid expansion is simultaneously reshaping demand, supply and financial markets, making it harder for central banks to assess changes in the economy.
The BIS estimates that the world’s five largest technology companies will invest more than $1 trillion in AI between 2025 and 2026. Industry forecasts indicate that worldwide AI investment could grow from roughly $500 billion currently to as much as $4 trillion by 2030.
Hernández de Cos said AI has real potential to boost productivity, especially in fields such as coding, consulting and professional writing. Studies referenced by the BIS have recorded productivity gains of between 10% and 65% for particular tasks.
However, those economic gains are not assured to be distributed evenly, while the sheer size of today’s investment surge presents risks that financial institutions and policymakers will have to monitor.
Financing is one area of concern. Hernández de Cos said a growing share of AI investment is being backed by debt and private credit instead of corporate earnings. He noted that much of this financing is difficult to evaluate because it involves interconnected and opaque structures.
Those vulnerabilities could become more significant if corporate earnings fall short of expectations or valuations of AI-related companies decline sharply.
Market concentration presents another risk. A relatively small group of technology companies is responsible for a significant share of the investment powering the AI build-out, potentially magnifying the financial impact of a downturn in the sector.
The technology is also changing patterns of international trade. Economies closely integrated into the AI supply chain, including South Korea, Singapore, Malaysia and Taiwan, have benefited from increased export prices for AI chips and related equipment.
India has likewise become an important beneficiary, supported by its digital infrastructure. Hernández de Cos said the country has a genuine chance to reduce its economic gap with more advanced economies.
AI could also produce major changes in employment. Although job losses have so far remained limited, signs of displacement are appearing in sectors such as customer service, programming and administrative work.
As businesses increase their use of AI systems, retraining and reskilling workers will therefore become increasingly important.
The BIS estimates that AI could lift total factor productivity growth by approximately half a percentage point each year. The effect will depend on the pace at which businesses adopt the technology and how successfully labor and capital are redirected toward productive activities.
For central banks, the task will be to weigh those potential productivity gains against vulnerabilities associated with elevated valuations, concentrated markets, debt-backed investment and shifting employment patterns.
Hernández de Cos drew comparisons between parts of the current investment surge and earlier periods of rapid technological development, including the railway boom and the dotcom era, while emphasizing that the eventual outcome remains uncertain.
Original article
Comments