The Head of the Bank for International Settlements (BIS), Mr Pablo Hernández de Cos, has warned that the rapid rise of artificial intelligence (AI) is creating new financial stability risks.
According to a BBC report, Hernández de Cos made the comments in London while speaking at a conference hosted by the Reserve Bank of India.
The BIS is an international financial institution owned by its member central banks. Founded in 1930, it serves as a bank for central banks. Its members include Brazil, Bulgaria, Canada, Chile, China, India, France, Germany, Italy and Japan.
He said spending on AI-related infrastructure was already at a scale significant enough to influence global economic conditions.
According to him, AI does not change the mandates of central banks, but it makes economies harder to interpret because it affects demand, supply and financial markets simultaneously.
He disclosed that the BIS estimates the world’s five largest technology firms will invest more than one trillion dollars in AI between 2025 and 2026.
De Cos also said industry forecasts suggest global AI investment could grow from about 500 billion dollars currently to as much as four trillion dollars by 2030.
“The promise of AI is real, but its long-term impact will depend on policy choices, investment in skills and infrastructure, and how widely benefits are shared,” he said.
The BIS chief said the AI boom was increasingly being financed through debt and private credit rather than corporate earnings. He said this required close scrutiny because much of the funding remained “opaque and interconnected.”
He added that AI was also changing global trade flows, with economies closely tied to the technology supply chain — including South Korea, Singapore, Malaysia and Taiwan — benefiting from stronger export prices for AI chips and equipment.
De Cos pointed to evidence that generative AI could significantly boost productivity, with studies showing gains of between 10 per cent and 65 per cent in specific tasks, particularly in coding, consulting and professional writing.
He said the broader question was how much of those improvements would translate into economy-wide productivity growth. Current estimates suggested AI could raise total factor productivity growth by about half a percentage point a year, depending on the pace of adoption and how effectively labour and capital are reallocated.
De Cos further said advanced economies were expected to benefit first because of their larger service sectors and greater readiness to deploy AI. Conversely, emerging economies faced more varied prospects, although India had a “genuine opportunity” to narrow the gap, helped by its digital public infrastructure.
He said that while AI could enhance workers’ productivity, it could also replace routine cognitive tasks. Job losses so far remained limited, but signs were emerging in customer service, programming and administrative roles.
He warned that lofty valuations, market concentration and opaque financing structures could create vulnerabilities if corporate profits fall short of expectations.
“I do not say that this is where the AI boom must lead, but the scale and speed of the current investment boom, and the weight of expected commercial returns, do warrant some caution,” he said.
