Key facts
- AI's rapid rise is creating new financial stability risks, according to Bank for International Settlements head Pablo Hernandez de Cos.
- The world's five largest technology firms are expected to invest over $1 trillion in AI between 2025 and 2026.
- Global AI investment could grow from $500 billion to as much as $4 trillion by 2030.
- AI-driven productivity gains of 10% to 65% have been observed in specific tasks like coding and professional writing.
- AI could raise total factor productivity growth by about half a percentage point annually.
- AI boom is increasingly financed through debt and private credit, which Hernandez de Cos said merits close scrutiny.
Pablo Hernandez de Cos, head of the Bank for International Settlements (BIS), has warned that the rapid ascent of artificial intelligence presents new risks to financial stability. He noted that substantial investments in AI infrastructure, already influencing global economic conditions, are increasingly being funded through debt and private credit, a trend that requires close monitoring due to its opaque and interconnected nature.
Central banks, while maintaining their monetary policy mandates, face challenges in interpreting economies as AI simultaneously impacts demand, supply, and financial markets. The BIS estimates that the five largest technology firms globally will collectively invest over $1 trillion in AI between 2025 and 2026. Industry forecasts suggest that global AI investment could surge from approximately $500 billion currently to as much as $4 trillion by 2030.
Hernandez de Cos acknowledged the real promise of AI but stressed that its long-term effects are contingent on policy decisions, investments in skills and infrastructure, and equitable benefit sharing. He pointed to evidence of significant productivity gains, ranging from 10% to 65% in specific tasks like coding and professional writing, due to generative AI. Economists estimate AI could boost total factor productivity growth by about half a percentage point annually, depending on adoption rates and efficient reallocation of labor and capital.
Advanced economies are expected to be the initial beneficiaries due to their larger service sectors and readiness to deploy AI. Emerging economies present varied prospects, though India has a notable opportunity to reduce the technological gap, supported by its digital public infrastructure. However, AI also poses risks, including potential job displacement in routine cognitive tasks, although current job losses are limited. Hernandez de Cos also highlighted vulnerabilities arising from high valuations, market concentration, and opaque financing structures, particularly if corporate profits fail to meet expectations. He drew parallels to past investment booms, such as the railway expansion and the dotcom bubble, to underscore the need for caution.