The CEO of one in every of China’s prime chipmakers simply despatched a stark warning in opposition to huge tech’s unprecedented race to construct out as many AI information facilities as it could possibly.
“Firms would like to construct 10 years’ price of knowledge heart capability inside one or two years,” Semiconductor Manufacturing Worldwide Corp.’s Co-CEO Zhao Haijun stated Wednesday on a name with analysts, per Bloomberg. “As for what precisely these information facilities will do, that hasn’t been totally thought by way of.”
Simply Alphabet, Microsoft, Meta, and Amazon alone are anticipated to spend almost $700 billion on AI this yr, repeatedly assuring buyers that demand outpaces provide. Over the subsequent three years, the worth tag of the info heart buildout is predicted to exceed $3 trillion.
These monetary commitments have began to spook buyers. In the meantime, the image that’s steadily forming has led to the present AI hype being in comparison with the dot-com bubble.
Invigorated by the surplus funding flowing into constructing out the web and anticipated demand for web providers, the telecommunications trade spent billions of {dollars} to put out fiber optic cables within the late Nineteen Nineties. By 2002, the dot-com bubble had lengthy burst, and nonetheless lower than 5% of the fiber optic community was reportedly in use. A telecoms crash rapidly adopted the dot-com bubble burst, and these unused fiber networks, referred to as “darkish fiber,” sat idle for years.
Ultimately, although, the demand for the web did pan out as anticipated, simply years late. As demand for the web grew exponentially, a variety of the infrastructure that was constructed out within the late 90s was ultimately used.
For AI, although, the scenario is likely to be a bit totally different: AI chips utilized in these information facilities have a comparatively clear expiration date, and if demand doesn’t pan out as anticipated earlier than these chips expire, then meaning these corporations may have flushed an entire lot of cash and sources down the drain.
Meta says its chips at the moment are good for roughly 5 and a half years, up from a earlier estimate of 4 years. Nvidia executives have claimed that the corporate’s chips that have been shipped out six years in the past are nonetheless in full use.
However even when the chips themselves are effective, their worth additionally depreciates as new, higher fashions get churned out. At its present fee, Nvidia releases a new flagship AI chip yearly.
The depreciation of the worth of chips is factored into firm earnings, however trade specialists are having a tough time seeing eye to eye on whether or not the present estimates are real looking.
Some argue {that a} six-year depreciation cycle is completely affordable, and older GPUs will nonetheless be fascinating as cheaper options when newer, extra superior variations come out. Then again, you’ve gotten buyers like Michael Burry (of “The Huge Quick” fame) who declare that the precise helpful lifetime of an AI chip is not more than 2-3 years.
“By my estimates, they may understate depreciation by $176 billion 2026-2028,” Burry stated in a put up on X.
In its newest annual report, Microsoft stated its “laptop gear” had an estimated helpful life ranging anyplace from two to 6 years.
