💥 AI MANIA: WHEN DREAMS OUTPACE CALCULATION

🔺 The artificial intelligence boom has morphed into a construction frenzy: tech giants are pouring hundreds of billions of dollars into data centers, chips, and energy infrastructure. The scale is staggering — in just three years, investments have exceeded the entire cost of building the U.S. interstate highway system. A striking example is Ellendale, North Dakota, where amid two motels and a Dollar General store, a $15B AI plant is rising — equal to a quarter of the state’s GDP.

🔺 Optimists compare this moment to the Industrial Revolution, but the core problem is obvious: AI revenue remains minuscule. OpenAI, despite contracts worth tens of billions, will generate around $13B this year. Microsoft, Meta, Amazon, and Alphabet plan to invest nearly $400B by 2025 — more than the Apollo program. Yet most AI chips last only 3–5 years, and demand is limited: corporate clients resist paying more than $30 per month per employee.

🔺 Historical parallels are unsettling. In the late 1990s, telecoms poured hundreds of billions into fiber-optic networks, betting on an internet explosion. The result — “dark fiber,” bankruptcies, and decades of underutilized capacity. Today’s data centers risk repeating that story, especially for intermediaries like CoreWeave, burdened with debt and long-term obligations against short-term contracts.

🔺 Still, the race is accelerating. Projects branded with names like “Stargate” or “Hyperion” are meant to highlight the grand ambitions. CoreWeave and its partners are erecting hangars powered by hundreds of megawatts, hoarding Nvidia chips and renting out capacity. Their soaring valuations astonish, yet much of it is fueled by high-interest debt.

🔺 Economists warn: to break even, the world would need $2T in annual AI revenues by 2030 — more than the combined earnings of all FAANG+Nvidia today. Reality is far from that mark: last year, sector revenue was just $45B. Studies from MIT and the University of Chicago show that 95% of companies see little measurable return from AI, while new models grow ever more expensive without delivering the promised breakthrough.

🔺 AI supporters counter that the differences from the dot-com era are crucial: tech giants generate immense cash flows, and AI is instantly accessible to hundreds of millions of users. ChatGPT already has more than 700M weekly users, and executives argue that AI could add 10% to global GDP.

🔺 But the risk of overheating is real. The history of the 19th-century “railway mania” and the collapse of the fiber-optic boom suggests that collective optimism often ends in painful corrections. AI will undoubtedly reshape the economy. The only question is whether the real payoff will justify today’s massive spending — or whether this boom will become yet another reminder that dreams of innovation often outpace rational calculation.

Content Hub - Articles Index - Library / Archive

#crisis

Comments

Post a Comment

Popular posts from this blog

🗂 INDEX - REVERSALS & TRENDS

USEFUL STUFF -   Crisis Map 2024-2030 - Long Term Cycles of the US Stock Market -  7-year Crisis Cycle of the US Stock Market -  Medium-Term US Stock Market Cycle -  New 4-year Cycle of the US Stock Market -  50-week Cycle of US Stock Market -  Market Crashes and US Presidents -  The $50 Trillion Bubble - Long Term Gold Cycles -  The Dollar's Cyclical Axiom -  Bear 2025 in Numbers (2.0) - Extremes and Pivots - The Long Cycle of the Yen -  Dollar-Yen 34-year High - Jupiter and Uranus in Crude Oil Cycles -  Long Crude Oil Cycle -  Crude Oil Astrology (Part 1) -  Crude Oil Astrology (Part 2) - Peak of Market Cycles. The End of Gemini Chaos. - A Few Words about Retro-Mercury -  Retrograde Mercury 2024 - Retrograde Mercury 2025 - Retrograde Venus - Retrograde Venus 2025 in Numbers - Mars in Scorpio - Reset Cycle - Russian Roulette -  Illusions of Gold Bugs - Benner Cycles - In Company with a Demon - End of Taurus Euph...
Read more

📈 EXTREMES & PIVOTS

Image
👉 Extreme forecast and pivot forecast are proprietary terms I use in my work to describe time-based projections of potential market reversals. 👉 These forecasts are designed to support swing or position trading – to help open or close trades. The forecasted reversal date is based on astrological data, while confirmation (and actual trade execution) comes from cyclical and technical analysis as the date approaches. ☝️ Not every projected extreme or pivot will play out! It's not known in advance whether the reversal will mark a top or a bottom. For swing and position trading in futures, that’s not critical – what matters is the trend that follows. 👉 An extreme or pivot may hit within a margin of a couple of days. Occasionally, instead of triggering a reversal, the level breaks and confirms continuation. Extreme and pivot forecasts for the stock market and gold are usually synchronized, and may sometimes apply to crude oil as well. ☝️ As the projected date approaches, technic...
Read more

📈 LONG-TERM CYCLES OF THE U.S. STOCK MARKET (Update)

Image
☝️ Historical retrospectives show that the longest cycle of the U.S. stock market spans approximately 72 years . In recent history, there have been only three such supercycles . 1️⃣ The first supercycle ended with the Panic of 1857 , which gradually transitioned into the Long Depression of 1873–1879 . 2️⃣ The second supercycle ended with the Great Depression of 1929–1932 . 3️⃣ The third supercycle ended with the Great Recession – the financial crisis of 2007–2008 . 👉 Technically, the third supercycle ended in March 2009 on a hard Saturn–Uranus aspect , whose oppositions and conjunctions have a 100% correlation with long-term U.S. market cycles. We are currently at the beginning of the fourth supercycle . 👉 Each supercycle consists of two long-term cycles of approximately 36 years . These 36-year cycles are further divided into two phases of 18 years . Historically, most 18-year cycles fall in a range from 15 to 21 years . Given that 2024 is the 15th year of the current 1...
Read more