Historical capital bubbles all have these five major rules. At which stage is AI currently?

The BCA Research report warns that the AI bubble will burst within 6 to 12 months, and recommends that investors pay attention to forward-looking indicators such as analyst expectations and GPU costs in the short term and medium term. This article is derived from an article written by Wall Street News and is compiled, compiled and contributed by Foresight News. (Synopsis: Huida Q3 earnings report smashes AI bubble theory, Huang Jenxun: Blackwell demand is booming, Nvidia pulls up 5% after hours) (Background supplement: “Big Sell” Michael Burry liquidated his holdings this quarter! Why turn to short Nvidia and Chinese tech stocks? From 19th-century railroads to 21st-century artificial intelligence, every major technological innovation in history has spawned a capital spending boom, but the frenzy has often ended in a bubble bursting. BCA Research's special report, “When Capex Booms Turn Into Busts: Lessons From History,” released in November, reviewed four typical capital spending booms, revealing the core logic of moving from boom to collapse and warning of the current AI boom. The report summarizes five common patterns: investors ignore the S-curve used by technology, revenue forecasts underestimate the decline in prices, debt becomes the core dependence on financing, asset price peaks precede investment declines, and capital spending collapse and recession reinforce each other. And these laws have shown signs in the current AI field: technology adoption has stopped, token prices have plummeted by more than 99%, corporate debt has surged, and GPU leasing costs have fallen. Based on historical control analysis, BCA Research concludes that the AI boom is following the path of the historical bubble and is expected to end in the next 6 to 12 months. The report recommends that investors maintain a neutral allocation to equities in the short term, moderately underweight stocks in the medium term, and closely track forward-looking indicators such as analyst expectation revisions, GPU leasing costs, and corporate free cash flow. In particular, the report pointed out that the current economic environment has added to the concern, and job vacancies in the United States have fallen to a five-year low. If the AI craze recedes and no new bubble offsets the impact, future recessions could be worse than they were when the Internet bubble burst in 2001. Historical Mirror: The Collapse Trajectory of the Four Capital Carnivals BCA said that the essence of the capital expenditure boom is capital's collective optimistic expectation of the commercialization prospects of new technologies, but history has repeatedly proved that this optimism often deviates from the objective law of technology landing, and eventually collapses in the imbalance between supply and demand, debt accumulation and valuation inflated. The Anglo-American railroad boom of the 19th century demonstrated the destructive power of overcapacity. According to the report, the success of the Liverpool-Manchester Railway in 1830 ignited a British investment frenzy, and railway stock prices nearly doubled between 1843 and 1845. By 1847, spending on railway construction as a percentage of British GDP soared to a record 7%. Tighter liquidity eventually triggered the financial crisis of October 1847, when the railroad index plunged 65% from its peak. The railroad boom in the United States culminated in the 1873 panic, the New York Stock Exchange was forced to close for ten days, and corporate bond default losses reached 36% of face value between 1873 and 1875, the report said. After U.S. railroads peaked at more than 13,000 miles in 1887, overcapacity led to a collapse in transportation prices, and by 1894 about 20 percent of U.S. railroad miles were in bankruptcy. The electrification boom of the 2020s exposed the fragility of the pyramidal capital structure. The report notes that the proportion of households with electricity climbed from 8% in 1907 to 68% in 1930, but the process was concentrated in cities. Wall Street was deeply involved in the boom, and utility stocks and bonds were advertised as safe assets “for widows and orphans to invest,” and by 1929 holding companies controlled more than 80 percent of electricity generation in the United States. Insul, the largest utility, went bankrupt in 1932 after the stock market crash of 1929, allegedly wiping out the life savings of 600,000 small investors, the report said. After peaking at about $919 million in 1930, U.S. spending on electric utility construction plummeted to $129 million in 1933. The Internet boom of the late '90s proved that innovation doesn't mean profit. According to the BCA, between 1995 and 2004, U.S. nonfarm productivity grew at an annualized rate of 3.1 percent, far outpacing subsequent periods. But technology-related capital spending soared from 2.9% of GDP in 1992 to 4.5% in 2000, and overinvestment put significant pressure on corporate balance sheets. The report notes that free cash flow in the telecommunications industry peaked in late 1997 and then declined sharply in 2000. After rising sixfold between 1995 and 2000, the Nasdaq Composite plunged 78% in the next two and a half years. Multiple oil booms perfectly illustrate the cycle of supply-demand imbalances. After the discovery of huge oil reserves in eastern Texas in 1930, production exceeded 300,000 barrels per day in 12 months, but the intensification of the Great Depression caused oil prices to plummet to 10 cents per barrel, BCA said. Saudi Arabia's abandonment of production restrictions in 1985 caused oil prices to fall to $10 per barrel. Between 2008 and 2015, the U.S. shale boom boosted crude oil production from 5 million barrels per day to 9.4 million barrels per day, while OPEC's refusal to cut production in 2014 sent oil prices down from $115 a barrel in the middle of the year to $57 at the end of the year. Five common laws: the only way from prosperity to collapse Reviewing the rise and fall of the four typical booms, BCA Research has summarized five common laws, which provide a key yardstick for judging the current trend of the AI boom. Specifically: The first rule is that investors ignore the S-shaped curve adopted by technology. Technology adoption is never linear, but follows an S-shaped curve of “early adopter acceptance – mass adoption – lagging group follow-up”. Stock prices typically rise in the first phase and peak in the middle of the second phase, when adoption growth turns from positive to negative. The AI landscape is showing this today: most companies have expressed an intention to increase their use of AI, but the actual adoption rate has shown signs of stalling, and some indicators have even declined in recent months. This divergence of “will and action” is a typical signal of the late stage of technology adoption. The second rule is that revenue forecasts underestimate the price decline. New technologies initially have pricing power due to scarcity, but as technology becomes more widespread and competition intensifies, prices are bound to drop significantly. Between 1998 and 2015, Internet traffic grew at an annualized rate of 67%, but the price per unit of information transmitted fell sharply. The price of solar panels has continued to fall since its inception, falling by 95% since 2007 alone. The AI industry is repeating the mistakes of the past: since 2023, the introduction of faster chips and better algorithms has caused the price of tokens to fall by more than 99%. Despite the emergence of new apps such as video generation, users' willingness to pay for such apps is unclear. The third rule is that debt has become the core dependence of financing. Companies in the early stages of the boom can usually meet capital expenditure needs through retained earnings, but as they invest…