Key Takeaways
- • Bridgewater Co-CIO Greg Jensen argues the AI bubble lies ahead, not behind, stating that the greatest bubbles require both a transformational technology and a favorable macro/monetary environment—conditions he believes are currently in place.
- • Investors significantly underestimate AI's scale and transformational potential, with Jensen warning that "investors have no idea what's hitting them" as AI spending accelerates into what he characterizes as a "dangerous" new phase.
- • Market skepticism about AI fundamentals persists despite strong execution, with doubts focused on: full valuation pricing, delayed LLM revenue translation, technical limitations in model outputs, slowing marginal improvements, and historical bubble parallels (Dot-Com comparisons).
- • AI infrastructure buildout is entering its fourth year with massive capital deployment, requiring an estimated $2 trillion in annual AI revenue by 2030 to justify current investment levels, according to Bain consulting estimates referenced in market commentary.
- • Nvidia's recent earnings disappointment signals incomplete market recognition, as record-beating results failed to generate a rally, suggesting the stock/theme may reflect investor skepticism rather than full pricing of AI's long-term potential.
Summary
Bridgewater Associates Co-Chief Investment Officer Greg Jensen has signaled that the artificial intelligence investment cycle remains in its infancy, with the most significant market disruption and potential bubble dynamics still ahead. Speaking on the "In Good Company" podcast in late November 2025, Jensen explicitly stated "the bubble is ahead of us, not behind us," challenging the prevailing market narrative that AI enthusiasm may have peaked. His thesis rests on a framework identifying the two necessary conditions for the greatest financial bubbles: a truly transformational technology and a supportive macroeconomic and monetary environment, both of which Jensen contends are currently present.
Jensen's assessment directly addresses a growing body of market skepticism that has emerged following Nvidia's recent quarterly earnings report. Despite the chipmaker posting results that beat analyst estimates on record, the stock failed to rally on the news—a signal Jensen interprets as evidence that the broader market has not yet fully internalized the scale of AI's transformational potential. The recent disappointment around Nvidia underscores investor doubts that have begun creeping into markets, including concerns that current valuations already reflect future AI upside, that large language models have not yet translated capabilities into meaningful revenue streams, and that technical progress is decelerating relative to mounting computational costs.
Jensen characterizes the current investment phase as "dangerous" and warns that investors fundamentally misunderstand the magnitude of what is unfolding. The AI infrastructure buildout is entering its fourth year, with capital deployment accelerating substantially. Bain & Company has estimated that sustaining the current wave of AI infrastructure spending would require $2 trillion in annual AI revenue by 2030—a figure that underscores the scale of investment relative to current monetization. This massive commitment of capital, combined with the nascent state of revenue realization from large language models and other AI applications, creates the structural conditions Jensen identifies as characteristic of major bubbles.
The market debate reflects legitimate technical concerns alongside macroeconomic positioning. Skeptics point to issues including LLM performance inconsistencies on multistep problems, diminishing returns on model improvements despite exponentially higher training costs, and the historical precedent of the Dot-Com bubble. Conversely, some institutional observers argue that contemporary AI investment differs fundamentally from past cycles due to the involvement of established institutional capital and foreign investors rather than retail speculation alone. Nevertheless, Jensen's warning suggests that even sophisticated market participants may be systematically underestimating the capital allocation shifts and volatility that could accompany full recognition of AI's transformational scope.
Jensen's assessment directly addresses a growing body of market skepticism that has emerged following Nvidia's recent quarterly earnings report. Despite the chipmaker posting results that beat analyst estimates on record, the stock failed to rally on the news—a signal Jensen interprets as evidence that the broader market has not yet fully internalized the scale of AI's transformational potential. The recent disappointment around Nvidia underscores investor doubts that have begun creeping into markets, including concerns that current valuations already reflect future AI upside, that large language models have not yet translated capabilities into meaningful revenue streams, and that technical progress is decelerating relative to mounting computational costs.
Jensen characterizes the current investment phase as "dangerous" and warns that investors fundamentally misunderstand the magnitude of what is unfolding. The AI infrastructure buildout is entering its fourth year, with capital deployment accelerating substantially. Bain & Company has estimated that sustaining the current wave of AI infrastructure spending would require $2 trillion in annual AI revenue by 2030—a figure that underscores the scale of investment relative to current monetization. This massive commitment of capital, combined with the nascent state of revenue realization from large language models and other AI applications, creates the structural conditions Jensen identifies as characteristic of major bubbles.
The market debate reflects legitimate technical concerns alongside macroeconomic positioning. Skeptics point to issues including LLM performance inconsistencies on multistep problems, diminishing returns on model improvements despite exponentially higher training costs, and the historical precedent of the Dot-Com bubble. Conversely, some institutional observers argue that contemporary AI investment differs fundamentally from past cycles due to the involvement of established institutional capital and foreign investors rather than retail speculation alone. Nevertheless, Jensen's warning suggests that even sophisticated market participants may be systematically underestimating the capital allocation shifts and volatility that could accompany full recognition of AI's transformational scope.
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.