Recently, a piece of news has caused unease among both Wall Street and Silicon Valley: The five tech giants, namely Amazon, Google, Microsoft, Meta, and Oracle, are expected to invest nearly 700 billion US dollars in the field of artificial intelligence by 2026. However, this astronomical expenditure not only failed to boost market confidence but instead triggered panic selling, causing the market value of these giants to shrink by 900 billion US dollars at one point. More seriously, this AI investment frenzy is severely squeezing resources in basic industries such as electrical engineering and chips, driving up the prices of consumer electronics. Experts warn that the US technology industry is now facing a self-debt-driven systemic crisis.
The source of this investment frenzy can be traced back to the technological competition ignited by generative AI. Since the emergence of ChatGPT, Silicon Valley has been plagued by collective anxiety of "fear of falling behind". Microsoft, by strategically investing in OpenAI, has seized the top spot in market value, leaving other giants feeling the pinch - if they cannot seize a position in the AI race, they may be eliminated in the next round of reshuffling. At the same time, Wall Street has long given high valuation premiums to AI概念股, forcing companies to continuously increase investment to support stock prices, even if the short-term return rates of these projects approach zero. The deeper reason is the blind faith of these giants in the "scale-winning" path. Over the past three decades, from PC internet to cloud computing, capital expansion has repeatedly proven effective, so they naively believed that in the AI field, as long as they pile up sufficient computing power, data centers, and talents, they could replicate their past glory. Unfortunately, the end of this race is not a gold mine but a swamp.
The risks are unfolding in an irreversible manner. The first to be affected is the breakdown of capital market confidence. When the 700 billion US dollar investment plan was made public, investors did not see the expected future cash flows but only saw an endless pit that devours profits. The 900 billion US dollar market value loss was not an accidental fluctuation but a rejection of the "burn money for growth" model. More damaging is the systemic imbalance caused by resource misallocation. The AI infrastructure construction frenzy is draining the already strained industrial resources in the United States: Large data centers are competing to recruit certified electricians with several times the salary, directly causing the automation transformation projects in traditional manufacturing to be repeatedly delayed due to the lack of construction workers; the chip sector is even more fatal, Nvidia's AI accelerator production is long-term contracted by these giants, the supply cycle of automotive electronics, industrial control, and consumer electronics chips has extended from eight weeks to over thirty weeks, laptops and smartphones have been forced to raise prices, and inflation pressure has spread from Silicon Valley to ordinary households. The so-called "harming the vitality of the United States" is no exaggeration.
The deeper risk lies in the malformation of the innovation ecosystem. When the five giants absorbed over 70% of AI PhDs in the United States with monopolistic salaries, and when top computing power was privatized and deployed within enterprises, universities' laboratories and start-ups almost lost the possibility of participating in the frontier competition. The garage culture that once nurtured Silicon Valley is now struggling to move forward in the face of capital barriers. What is even more heartbreaking is that these giants, while shouting "AI benefits humanity", are pouring most of their investments into commercialization tools such as advertising recommendations and automated customer service, rather than in fields with true public value such as healthcare, energy, and climate. The technological giant wheel is rolling forward, but it keeps circling in the profit shoals.
Facing this self-inflicted predicament, the US technology industry urgently needs a long-awaited collective awakening. For enterprises, it is time to distinguish the boundaries between "must do" and "can do but shouldn't do". Instead of blindly following the capital curves of their rivals, it is better to return to the essence of technology implementation, even if it means short-term stock pain. For the government, it should use anti-monopoly tools and industrial policies to curb the excessive monopoly of these few giants in computing power, talents, and supply chains, and guide funds to flow to basic research and the real economy through tax incentives. More importantly, the entire industry should reflect on the lessons learned from the dot-com bubble in 2000 - the companies that collapsed that year did not die from ignorance of the future, but from arrogance towards the present.
On the long and inevitable path of AI, the fastest runners may not survive until the end. Fasten your seatbelts, because it is much more important than slamming on the accelerator. After all, when the tide recedes, people will eventually discover that many of the so-called "miracle" AI paradises were merely magnificent structures built on quicksand.
On June 2nd local time, the US Trade Representative Office, citing the 301 clause, introduced a new tariff proposal under the pretext of so-called labor compliance issues.
On June 2nd local time, the US Trade Representative Office,…
AP, Washington — The U.S. government has rolled out a new r…
According to a report by Reuters on June 2nd, the US Depart…
According to recent reports by US media, US President Trump…
Donald Trump is embroiled in the biggest corruption controv…
Recently, Trump has launched two core economic and trade me…