Key Takeaways
- • Shanghai-based Banxia Investment Management and Wealspring Asset Management have publicly warned that global AI-related equities have entered "super bubble" territory, with valuations increasingly detached from fundamental growth prospects.
- • Banxia identified slowing momentum at AI model developer Anthropic as a potential early warning signal that hyperscaler AI monetization expectations may be overly optimistic, urging investors to exercise "very, very" high levels of caution with AI exposure.
- • Wealspring argues that many AI infrastructure companies lack durable competitive advantages and remain heavily dependent on continuous capital investment, suggesting current valuations do not adequately reflect execution risks and long-term sector economics.
- • Despite hedge fund skepticism, AI-linked stocks—particularly semiconductor and memory names including Micron Technology and SK Hynix—have delivered outsized gains driven by data center spending expectations, creating a widening institutional investor divide over valuation sustainability.
- • The warnings highlight a growing divergence among institutional investors regarding whether current AI equity valuations can be justified by future cash flows, even as AI indices continue to outperform broader benchmarks in 2026.
Summary
A cohort of prominent Chinese hedge fund managers is publicly cautioning that the global artificial intelligence equity rally has become disconnected from underlying fundamentals and is approaching bubble conditions. According to Bloomberg, Shanghai-based Banxia Investment Management and Wealspring Asset Management have emerged as leading voices expressing concern about current AI sector valuations.
Wealspring Asset Management has characterized AI equities as a "super bubble," contending that numerous companies within the AI infrastructure supply chain are being valued on aggressive growth assumptions despite lacking sustainable competitive moats. The firm emphasized that many businesses across the AI value chain remain structurally dependent on continuous capital deployment to sustain expansion, raising questions about the durability of current business models.
Banxia Investment Management has identified specific early warning indicators of a potential correction. The firm pointed to moderating revenue growth expectations for leading AI developers and flagged decelerating momentum at Anthropic, an AI model developer, as a potential harbinger that investor expectations for hyperscaler AI monetization may prove unrealistic. Banxia's founder advised investors pursuing AI sector exposure to exercise "very, very" high levels of caution going forward.
The hedge fund concerns stand in contrast to the broad outperformance of AI-linked equities across global markets. Semiconductor and memory companies—including Micron Technology and SK Hynix—have posted substantial gains, primarily driven by expectations of sustained capital expenditure on data centers and AI infrastructure. However, institutional fund managers increasingly question whether these valuations appropriately incorporate execution risks and reflect the sector's long-term economic fundamentals.
The divergence in institutional investor sentiment underscores a fundamental debate within the investment community: whether current AI equity valuations can be sustained by future cash flows. While avoiding AI exposure has proven costly in the near term, with AI indices continuing to outperform broader equity benchmarks throughout 2026, the warnings from Chinese hedge funds illustrate growing skepticism regarding the sector's fundamental support for prevailing price levels.
Wealspring Asset Management has characterized AI equities as a "super bubble," contending that numerous companies within the AI infrastructure supply chain are being valued on aggressive growth assumptions despite lacking sustainable competitive moats. The firm emphasized that many businesses across the AI value chain remain structurally dependent on continuous capital deployment to sustain expansion, raising questions about the durability of current business models.
Banxia Investment Management has identified specific early warning indicators of a potential correction. The firm pointed to moderating revenue growth expectations for leading AI developers and flagged decelerating momentum at Anthropic, an AI model developer, as a potential harbinger that investor expectations for hyperscaler AI monetization may prove unrealistic. Banxia's founder advised investors pursuing AI sector exposure to exercise "very, very" high levels of caution going forward.
The hedge fund concerns stand in contrast to the broad outperformance of AI-linked equities across global markets. Semiconductor and memory companies—including Micron Technology and SK Hynix—have posted substantial gains, primarily driven by expectations of sustained capital expenditure on data centers and AI infrastructure. However, institutional fund managers increasingly question whether these valuations appropriately incorporate execution risks and reflect the sector's long-term economic fundamentals.
The divergence in institutional investor sentiment underscores a fundamental debate within the investment community: whether current AI equity valuations can be sustained by future cash flows. While avoiding AI exposure has proven costly in the near term, with AI indices continuing to outperform broader equity benchmarks throughout 2026, the warnings from Chinese hedge funds illustrate growing skepticism regarding the sector's fundamental support for prevailing price levels.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.