Overinvestment might cause economic crisis, it happened in history and happened over and over again.
Railway at the end of the 19th century
History is always surprisingly similar. After the Civil War, the three major American families frantically built railroads. In the five years from 1868 to 1873, 70,000 kilometers of railroads were built. The total mileage of China’s railways is now only 150,000 kilometers.
However, agriculture had not yet been fully developed at that time, and oil was only concentrated in a few areas. In many places, there was no demand for freight at all. The entire railway industry was unprofitable, resulting in serious financial losses. More than 60 railway companies went bankrupt overnight, and 18,000 people lost their jobs. Many companies went bankrupt.
This economic depression lasted for seven years and was the largest economic crisis in the Western world in the entire 19th century, second only to the Great Depression of the 1930s. Therefore, overinvesting is extremely dangerous.
Dot-com bubble of 2000
WorldCom in 2000 was one of the largest communications companies. It invested hundreds of billions of dollars to lay fiber optic backbone networks across the United States and acquired other communications companies. President Clinton at the time even visited the company to inspect. This is called the “information highway” of the United States and the greatest birthplace of the third industrial revolution.
At that time, people were very excited. Employees of WorldCom even invested all their wages in buying company stocks because they believed this was a great era. The company’s stock has soared, and employees are confident about the company’s future.
However, in June 2000, the Internet bubble burst. In just three months, WorldCom’s stock price fell from its peak to “zero” and eventually went bankrupt. The reason was that it believed too much that the Internet revolution would Quick success.
At that time, the charging model of the Internet was based on traffic. However, the traffic demand of the Internet was not large at that time. The main Internet applications were only some e-commerce and websites, so there was no return. This phenomenon did not occur until the emergence of smart phones and people’s extensive use of video and audio, which led to a surge in traffic and ultimately made the Internet business model profitable.
Regarding the dotcom bubble in 2000, for details, please see my post of “How crazy was dot-com bubble? Companies have rise like a phoenix from the ashes”
What about now?
Is current AI overinvestment?
Recently, more and more people are discussing whether AI will bubble? Some people think that the future of AI is undoubted, but another group of people are worried that the major technology giants’ excessive investment in AI may fall into a dilemma that is difficult to recover?
This is indeed a very important issue. Starting from 2022, investors have invested US$600 billion, but the more they invest, the bigger the hole becomes. This shows that AI has not yet been able to generate sufficient returns. Currently, in the field of AI There is no “killer” application yet.
Without a killer application scenario, it will not be able to attract the average person to use it. If the average person does not use it, AI will not be popularized. If the technology or product cannot be popularized, there will be no profitable business model, and of course it will not be able to generate a lot of profits.
AI now faces similar challenges, raising concerns about overinvestment. If investments are made too far ahead, companies that invest heavily in AI infrastructure may eventually face huge losses if they are unable to generate sufficient returns, just like WorldCom (later acquired by Verizon Communications, ticker: VZ). Same mistake.
Regarding the actual cost of artificial intelligence, please see my post of “How Much Does Generative AI Cost?” for details.
AI investment crowds out other industries
Venture capital is a perspective from which to observe this phenomenon. Although venture capital accounts for a small proportion of the total social investment in the United States, it has a very strong role in promoting the industrial development and economy of the United States.
In the U.S. venture capital category in 2023, the proportion of AI investment will rapidly increase from single digits in 2022 to 20%. At the same time, the total amount of venture capital investment in the United States did not increase in 2023, and even hit a new low since 2017. When the total amount of venture capital investment decreases, the proportion of AI investment increases again, which results in squeezing investment in other industries.
But the problem is that investment in AI has not improved the real economy of the United States. The AI industry currently improves existing processes by automating certain tasks or improving worker efficiency. It does not create new, large-scale business scenarios, so the benefits generated are limited.
Data from Sequoia in March this year showed that the sales of new generative artificial intelligence startups were only US$3 billion, less than 7% of the financing amount in 23 years.
There are no new business scenarios and limited revenue generation, which means that radical AI investment has not brought more jobs. In addition, AI’s squeeze on investment in other industries has also reduced jobs in other industries. The number of new non-agricultural jobs in the United States in a single month has dropped from 350,000 at the beginning of 2023 to 142,000 in August this year.
Tech behemoths’ view
The high-intensity investment did not bring returns. In the second quarter, AI-related revenue accounted for 8% of Microsoft Azure, which increased by 1pc from the previous quarter. Office 365, which was second in AI closeness, dropped by about 2pc from the previous quarter. There is still no obvious sign of acceleration in the revenue contributed by AI.
Regarding the disproportion between investment and return, Meta and Google generously admitted in the conference call that “it is very possible that they and their peers have overinvested in AI.”
Is high-intensity AI investment still necessary? For giants, missing the next trend is more serious than overinvesting. Google has made it clear that it is better to overinvest than underinvest. Under the phobia of missing out, the AI arms race among large technology companies may continue.
This kind of competition will also affect most companies. Under the temptation of the technological wave and the pressure of investors, many companies have to follow AI, even if AI cannot produce results. For example, Salesforce has invested heavily in AI, but after investing these funds, revenue has barely grown.
AI’s big gamble can only be verified by time.
Related articles
- “Economic crisis caused by overinvestment“
- “Why software underperforming amid the AI craze?“
- “How Much Does Generative AI Cost?”
- “Major artificial intelligence companies in US stocks market“
- “Artificial intelligence investment trap“
- “The artificial intelligence bubble in the capital market is forming“
- “How crazy was dot-com bubble? Companies have rise like a phoenix from the ashes”
Disclaimer
- The content of this site is the author’s personal opinions and is for reference only. I am not responsible for the correctness, opinions, and immediacy of the content and information of the article. Readers must make their own judgments.
- I shall not be liable for any damages or other legal liabilities for the direct or indirect losses caused by the readers’ direct or indirect reliance on and reference to the information on this site, or all the responsibilities arising therefrom, as a result of any investment behavior.
