November 3, 2026, (Inside AI) — A senior official at China's financial regulator warned on Tuesday that global capital is pouring into artificial intelligence so heavily that it is starving traditional industries of credit and inflating leverage across the sector. Xiao Yuanqi, vice head of the China National Financial Regulatory Administration, delivered the warning at the Hong Kong Monetary Authority's Global Financial Leaders' Summit, an event that draws central bankers and bank chiefs from across Asia and beyond.
The timing matters. AI valuations have climbed through 2026 on expectations of enterprise adoption and data center buildouts. Xiao's remarks place a major regulatory voice against that consensus, and they come from the world's second-largest economy, whose banks and local governments carry heavy exposure to property and manufacturing. His argument is not that AI is a bubble in the conventional sense. It is that the boom is redistributing credit away from productive sectors that still employ most workers.
Why Beijing Sees Crowding Out, Not Growth
Xiao described the mechanism in concrete terms. Capital floods into a handful of technology leaders, leverage inside those firms and their backers rises, and the expansion that follows can outrun fundamentals. Meanwhile, traditional manufacturers and service firms face higher borrowing costs because lenders now see AI as the safer or more lucrative bet.
Read: IMF Chief Warns AI Risks May Outweigh Opportunities Without Safeguards
"This concentration of capital creates a crowding-out effect on traditional industries ... it raises borrowing costs and increases the financial burden on these firms, thereby weakening their capacity for reinvestment and expanded production, while simultaneously heightening credit risks," Xiao Yuanqi, vice head of the China National Financial Regulatory Administration, said at the summit.
He also framed the surge as a leverage problem rather than a pure valuation problem. "A massive influx of capital into the technology sector and leading firms has driven up leverage levels within these areas -- potentially fueling excessively rapid expansion," Xiao Yuanqi said.
The distinction carries weight. Leverage concerns invite prudential tools such as capital requirements and lending caps. Valuation concerns invite market-based corrections. Xiao's language points toward the first approach, which would give regulators a formal hook to intervene in credit allocation without directly targeting AI companies.
His prescription was broad but pointed. "Policymakers should prepare policy frameworks specifically addressing this AI-driven transformation, in terms of macroeconomic management and the prevention of economic and financial risks," Xiao Yuanqi said.
The warning lands in a year when AI infrastructure spending has become a macro story in its own right. Hyperscalers and chipmakers have committed hundreds of billions of dollars to data centers, and much of that spending is financed through debt and complex vendor arrangements. When one sector absorbs that much capital, the cost of money for everyone else tends to rise. Xiao's crowding-out thesis is a familiar one in development economics, now applied to the AI cycle.
Not everyone shares the diagnosis. Many investors argue that AI spending reflects genuine productivity gains and that credit markets remain deep enough to fund both new and old industries. They note that technology booms have historically lowered costs across the economy once the infrastructure matures. Xiao did not dispute AI's long-term value. His concern was the transition period and who absorbs the strain while it unfolds.
The venue added diplomatic weight. Hong Kong has positioned itself as a bridge between mainland regulators and global capital, and the summit gave Xiao a platform before an audience of international bankers. His remarks signal that Beijing intends to treat AI finance as a supervisory matter, not only an industrial policy one. That stance could shape how Chinese banks underwrite technology exposure and how foreign lenders structure cross-border AI deals.
Read: South Korea's Lee Proposes Record Budget With $15.97 Billion for AI
For now, the speech is a warning rather than a rule. No specific measures were announced. But regulators rarely speak this directly about a single sector without follow-up work in progress. Bank supervisors in Beijing have spent the past two years tightening scrutiny of concentrated credit exposures, and AI lending fits neatly into that framework.
The broader question is whether other regulators follow. European and U.S. authorities have focused on AI safety and competition, not credit allocation. If China formally links AI investment to financial stability risk, it would become the first major economy to regulate the boom through the banking system rather than the technology sector. That would be a notable shift in the global policy conversation, and one worth watching as 2027 budget and lending plans take shape.
