Michael Burry warns Nvidia's power-hungry AI chips could hand China the edge in global AI race

Michael Burry Sounds Alarm on Nvidia's Energy-Intensive Strategy

Michael Burry, The Big Short investor who famously predicted the 2008 financial crisis, has issued a stark warning about Nvidia's dominance in artificial intelligence. In a series of posts on X over the weekend, Burry argued that Nvidia AI chips—increasingly powerful and energy-intensive—could inadvertently hand the AI edge to China in the global AI race1

1

. His critique centers on a fundamental constraint: America's aging power infrastructure cannot keep pace with the escalating electricity demands of advanced AI systems.

Burry responded to a user who described Nvidia as the "gangster of the AI neighborhood" for shutting down narratives that hint at reducing GPU demand, replying "Exactly and sadly"1

1

. He contends that Nvidia has framed AI progress as merely figuring out "how to power and to cool bigger, hotter silicon," effectively locking the industry into what he calls a power consumption roadmap2

2

.

China's Energy Generation Capacity Advantage Grows

Burry's most compelling evidence lies in the numbers. He shared data showing China possesses more than double the electric generation capacity of the United States, with its installed electricity generation capacity surging far beyond that of the US and Europe since the early 2000s1

1

2

2

. "It is not just the total power advantage," Burry wrote. "It is the slope"2

2

.

This gap could prove decisive as AI systems demand increasing amounts of electricity. Burry contrasted China's infrastructure push with the US energy grid's constraints, noting that "U.S. transmission grid development is actually decelerating due to permitting issues, while China is building transmission at will to match power output"2

2

. That mismatch leaves US companies "plowing capital into a race it is structurally positioned to lose"1

1

, given China's ability to rapidly build power sources supporting large-scale AI deployments and data centers.

Call for Energy-Efficient ASICs Over Power Hungry Chips

Burry advocates for a strategic pivot away from increasingly power hungry chips. "The U.S. needs to get away from bigger and bigger power-hungry chips and innovate with AI-tuned ASICs like nobody's business," he stated2

2

. Energy-efficient ASICs—application-specific integrated circuits designed to perform specific tasks more efficiently—could offer a more sustainable path forward. However, Burry claims Nvidia maintains a "death grip on development" through its deals with key players across the AI ecosystem1

1

2

2

.

He argues that efficiency improvements have failed to keep pace with the sheer growth in computing being deployed, pushing electricity needs ever higher2

2

. The US must shift away from brute-force approaches centered on massive GPUs and focus instead on more efficient designs that don't strain the transmission grid.

Nvidia's Market Dominance Amid Criticism

The critique arrives as Nvidia sits at the center of the AI boom. The company's stock has surged more than twelvefold since early 2023, achieving a market capitalization of approximately $4.40 trillion—making it the world's most valuable publicly traded company1

1

2

2

. Year-to-date, Nvidia has gained 30.86%, while over the past five years the stock surge has reached 1,293.30%2

2

.

In the first nine months of this year, Nvidia generated roughly $148 billion in revenue and $77 billion in net income1

1

. The company posted third-quarter revenue of $57.0 billion, representing a 62% increase year-over-year and exceeding Street consensus estimates of $54.88 billion2

2

. CEO Jensen Huang recently said demand for the company's Blackwell chips is soaring and cloud GPUs are effectively sold out1

1

.

Broader Skepticism About AI Valuation Bubble

Burry, who now writes on Substack after stepping away from running a hedge fund, has positioned himself as a skeptic of the AI rally. He has argued that Nvidia and other AI-related companies are inflating a historic tech bubble1

1

. Among his criticisms, Burry has raised valuation concerns by alleging that Nvidia's customers have overstated the useful life of chips to stretch depreciation schedules and boost short-term earnings.

He has also taken issue with what he describes as "give-and-take" arrangements between Nvidia and major customers such as OpenAI and Oracle, and has criticized Nvidia's stock-based compensation as excessive1

1

. Despite his sharp words, Burry suggested Nvidia represents a relatively small short position for him, writing: "I believe it will fall, and is a pure play on my overall thesis, but it is not the worst out there"1

1

.