news

한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina

Author: Zhou Yixiao
Email: [email protected]

Big drop. Big drop. It’s a big drop for Mag7.

On August 5, the U.S. stock market ushered in "Black Monday" on Monday's trading day. Among them, the technology giants collectively plummeted. Nvidia, Apple, Microsoft, Google, Amazon, Tesla and Meta, known as the Mag7 "Seven Sisters", wiped out $600 billion in market value overnight.

Data shows that at the close of the day, Apple fell 4.82%, Amazon fell 4.1%, Google fell 4.45%, and Meta fell 2.54%. Microsoft fell 3.27%, and its market value fell below $3 trillion. Most chip stocks also fell, with Intel falling 6.38% and Nvidia falling 6.36%. This is a continuation of the decline that began last week. According to statistics from the US media, these seven companies have lost $3 trillion in market value in the past month.

Prior to this, on July 16, the S&P 500 reached its all-time high: $47 trillion. Of the 500 listed companies, these seven alone account for a third of the market value. In other words, 1.4% of the companies are worth more than $16 trillion.

However, in the past month, at the height of the tech company earnings season, tech giants have failed to convincingly show Wall Street that their huge investments in AI are paying off - at least in the short term, and Wall Street has begun to question the increasing costs of purchasing high-end chips and building data centers, and the giants' spending on AI technology has far exceeded Wall Street expectations. Despite spending billions of dollars, large technology companies have still had little success in terms of significant revenue growth or profitable new products brought about by AI, and investors have begun to become anxious.

As panic broke out across the market, investors began to re-evaluate their investment strategies, and anxious Wall Street began to get tired of Silicon Valley's AI story.

1

Buffett "retreats early" and sells half of Apple shares

Every move Buffett makes is like a barometer of the market, and now the Oracle of Omaha is also selling technology stocks.

On the 3rd local time, Berkshire Hathaway, owned by the well-known American investor Buffett, released its second quarter financial report for 2024. The financial report showed that the company reduced its holdings of Apple shares from 789 million shares in the first quarter to about 400 million shares in the second quarter, a decrease of nearly 50%.

This is not the first time that Berkshire has reduced its holdings in Apple. Previously, Berkshire Hathaway had reduced its holdings in Apple by 13% in the first quarter. Buffett said at the company's shareholders meeting in May that the reduction in Apple's stock was due to tax reasons after the investment made considerable gains, rather than based on a long-term judgment on Apple's stock.

However, even after the reduction, Apple is still Berkshire's largest single holding. Some believe that more and more investors, including Berkshire, may be waiting to see whether Apple's AI investment can bring actual returns. At present, it seems that they are not satisfied with the speed of Apple's AI monetization.

It is worth noting that Buffett chose to sell Apple after Tim Cook released the "epoch-making" ChatGPT version of Siri at WWDC24. This timing may imply Buffett's judgment on Apple's AI strategy.

Jim Awa, senior managing director of Clearstead Advisors, said: "Buffett may feel that we are about to enter a recession, and he is cashing out now in order to buy high-quality companies at low prices in the future."

1

Nvidia once fell 15%, Huang Renxun himself sold $320 million in July

Nvidia, which has seen the most dramatic gains in this wave of AI madness, hit a record closing high of $135.58 on June 18, then experienced ups and downs, and finally began to decline in early July. It has now fallen 26% from its closing price on July 10. It also fell 15% during Monday's plunge.

Public information also shows that Nvidia's founder and CEO Huang Renxun sold a record $323 million of Nvidia stock in July. Before the recent plunge, Huang Renxun sold a total of about $500 million of Nvidia stock in the past two months.

Nvidia's plunge also has a direct "cause." According to The Information, Nvidia has informed Microsoft and at least one other cloud provider that the production time of the Blackwell B200 AI chip will be at least three months longer than originally planned.

The delay is due to a design flaw that was discovered unusually late in the production process. The B200 chip, a highly anticipated successor to the extremely popular and hard-to-get H100 chip, is now not expected to ship in large quantities until the first quarter of 2025. This delay could have a ripple effect on the entire AI industry chain.

This undoubtedly casts a shadow on the AI ​​layout of technology giants. In fact, major companies have already invested huge amounts of money in the field of AI. In the second quarter ending in June, capital expenditures of Google's parent company Alphabet, Amazon, Microsoft and Facebook's parent company Meta Platform totaled nearly $60 billion, a year-on-year increase of two-thirds. Most of the spending went to Nvidia.

The frenzy of investment has also caused concern among some investors. Elliott Management, a hedge fund that manages about $70 billion in assets, told investors in an open letter that Nvidia is in a "bubble." The letter concluded that artificial intelligence is essentially a software that has not yet provided "value commensurate with the hype" so far. If Nvidia reports poor financial results, it could cause the current bubble to burst, thus "breaking the spell."

Elliott Management's warning comes as chip stocks, which had surged on investor bets on the potential of generative artificial intelligence, are now falling on concerns about whether big companies will continue to invest heavily in AI. The sharp shift in market sentiment reflects investors' uncertainty about the prospects of the AI ​​industry.

Concerns seem to have begun to be confirmed in the market. After the release of its financial report last Thursday, Intel also plummeted 26% after the market, hitting its biggest drop in 50 years, and planned to lay off 15,000 employees. Other US chip stocks also fell.

1

Microsoft: AI is a bottomless pit of money, and 29% growth is not enough

Microsoft's fast-growing artificial intelligence business can't grow fast enough for Wall Street.

When a company's most important division grows 29% in the quarter, Wall Street should be cheering like an Olympic champion.

But last week, Microsoft announced that its Azure cloud business, the core of its AI strategy, had achieved 29% growth, and its stock price plummeted 7%. It is now more than 10% lower than a month ago. Investors are worried that this deceleration may indicate that the AI ​​boom is cooling down and future growth will slow down.

First, it suggests that the current AI mania has gotten out of control, with investors valuing stocks in a state of perfection, as if hypergrowth can continue forever. If a company’s performance falls short of expectations, at least in the eyes of Wall Street, its share price will suffer.

Of course, Microsoft executives explained this slowdown in part because of supply and demand. Building cloud infrastructure (data centers) takes time, which is why Microsoft is working with companies like Oracle to fill the gap.

Second, investors seem to think that AI profits are easy to come by and do not require much upfront investment, but this is not the case. Building these AI-friendly data centers and purchasing AI chips are very expensive. Tech giants spend tens of billions of dollars on data center construction and expansion every year, and there is no end in sight.

For example, Microsoft's capital expenditures in the most recent quarter reached $13.9 billion ($19 billion if leasing is included), a year-on-year increase of 55%. Microsoft executives expect to maintain this level of spending in the future, and this continued high-investment strategy has also raised market concerns about the return cycle. Last week, Google's parent company Alphabet also reported high infrastructure costs, resulting in a sharp drop in its stock price.

1

Wall Street can't wait

Wall Street has always been a booster of the new technology bubble. Whenever it smells the breath of innovation that may bring revenue growth, it will be excited. This time is no exception. Since ChatGPT, the potential of generative AI has been hyped. Wall Street investment bank analysts have continuously issued bullish reports, and fund managers are scrambling to adjust their asset portfolios to capture this wave of AI enthusiasm.

Last September, Goldman Sachs published an article titled "Why AI is not a bubble." It focused on the potential of artificial intelligence and believed that the prices of artificial intelligence-related stocks have not reached the typical bubble trading price. Goldman Sachs wrote: "We believe that this new technology cycle, which is expected to bring more outstanding performance, is still in a relatively early stage."

This year, Goldman Sachs' tone has changed to "maybe it's just a bubble." Their latest report is "Gen AI: too much spend, too little benefit?" Faced with the "change of face" of their comrades, CEOs of major technology companies have stated that everything is under control and emphasized the importance of AI to the future of their companies.

“When you go through an inflection point like this, the risk of underinvesting for us is much greater than the risk of overinvesting,” Google CEO Sundar Pichai told analysts.

However, these technology leaders also admit that what they need is time - a long time.

Microsoft's chief financial officer said on the company's earnings call that it expects its data center investments to support the monetization of AI technology "over the next 15 years and beyond." Meta expects that "the returns of generative AI will be realized over a longer period of time" and "we are still in the very early stages of generative AI... We do not expect generative AI products to be a significant revenue driver in 2024."

However, some investors have expressed doubts about this, believing that "if the return can only be obtained after 10 to 15 years, it is venture capital, not investment by listed companies." Not long ago, they were optimistic about the long-term prospects of artificial intelligence.

Since ChatGPT set off the AI ​​arms race 18 months ago, tech giants have promised that the technology will revolutionize every industry, justifying spending tens of billions of dollars on data centers and semiconductors needed to run large AI models. Compared to this vision, their current products seem puny: chatbots with no clear path to monetization, cost-saving measures such as AI programming and customer service, and AI searches that sometimes make up information.

Clearly, ChatGPT proved that the technology is feasible. But AI hasn’t really delivered results commensurate with the resources it has, which has led to a shift in tone. “There’s not an application that’s cost-effective right now,” Jim Covello, a well-known Goldman Sachs analyst, said in a company podcast.

Everything is pointing to the bursting of a bubble. Wall Street, the biggest advocate of AI yesterday, has now become the biggest skeptic and the needle that bursts the bubble it helped create.

Click "" and go