BIS Chief Warns AI Boom Creates Financial Stability Risks

AI Investment Surge Reaches Macroeconomic Scale

The artificial intelligence boom has grown large enough to influence global economic conditions and create new financial stability risks, according to Pablo Hernandez de Cos, head of the Bank for International Settlements. Speaking at a conference hosted by India's central bank, Hernandez de Cos highlighted how the scale and financing structure of AI investments now warrant careful monitoring by financial regulators.

The BIS estimates the world's five largest technology companies will invest more than $1 trillion in AI infrastructure between 2025 and 2026 alone. Industry forecasts project total global AI investment could surge from approximately $500 billion today to as much as $4 trillion by 2030, according to details first reported by Reuters.

Why it matters

When a single technology sector commands investment flows measured in trillions of dollars, it stops being just an innovation story and becomes a systemic economic force. Central banks now face the challenge of interpreting economies where AI simultaneously affects demand, supply, and financial markets—complicating traditional monetary policy analysis even as core mandates remain unchanged.

Debt Financing Raises Red Flags

A particular concern centers on how this investment wave is being funded. Hernandez de Cos noted the AI boom is increasingly financed through debt and private credit rather than corporate earnings, creating financing structures that remain "opaque and interconnected." This lack of transparency merits close scrutiny from financial stability authorities.

The BIS chief also pointed to vulnerabilities from lofty valuations, market concentration, and the risk that corporate profits could fall short of elevated expectations. While acknowledging AI's genuine promise, he drew parallels to historical investment booms including the railway expansion era and the dotcom surge.

Productivity Gains and Economic Divergence

Research shows generative AI can deliver significant productivity improvements for specific tasks, with studies finding gains between 10% and 65% in areas like coding, consulting, and professional writing. Current estimates suggest AI could boost total factor productivity growth by roughly half a percentage point annually, though this depends heavily on adoption rates and how effectively economies reallocate labor and capital.

Advanced economies with larger service sectors and greater technological readiness are expected to benefit first. Emerging economies face more varied prospects, though Hernandez de Cos said India has a "genuine opportunity" to narrow the gap, supported by its digital public infrastructure.

Trade and Labor Market Effects

The AI supply chain is already reshaping global trade flows. Economies including South Korea, Singapore, Malaysia, and Taiwan have seen stronger export prices for AI chips and equipment as they supply the infrastructure buildout.

On employment, while AI can enhance worker productivity, it also threatens routine cognitive tasks. Job losses remain limited so far, but early signs are emerging in customer service, programming, and administrative roles—making workforce retraining increasingly critical.

Reuters first reported these remarks from the BIS conference in India.