Stock mrket analysts and investment bankers have been saying for a few months the the AI bubble rapid i.e.the surge in the value of these companies based on stock markey trading in their shares, about to burst? Big tech’s gargantuan capital outlays – surging tenfold in the three years since ChatGPT was publicly launched – look increasingly unsustainable.
By some measures, artificial intelligence (AI) now sucks up more than half of America’s combined investment pool. Yet the promised uplift to business productivity hasn’t materialized.
According to techopedia.com, investment in artificial intelligence has sailed past levels last seen during the peak of the dotcom bubble. Jens Nordvig, founder and CEO of macroeconomic consultancy Exante, says AI-related capital expenditures now represent about 2% of America’s GDP. But scepticism is in short supple with people like Larry Fink, CEO of Blackrock, the world's biggest asset management firm fully on board with the hype.
Big tech AI data center spending has risen 10x in the last three years alone. The arrival of ChatGPT in 2022 has led to an AI bubble so big it’s consuming more than half of America’s cumulative investment pool.
So unquenchable is the industry’s thirst for cash that the world’s biggest economy may not have enough to slake it. OpenAI chief Sam Altman has already been fundraising beyond America’s shores, hoping to raise some $7 trillion from sovereign wealth funds and other sources in the UAE and elsewhere.
The AI euphoria driving this is partly due to global debt having ballooned from $64 trillion in 2000 to $338 trillion today.
Every major asset — stocks, bonds, property — has been inflated by
decades of easy money. What looks like prosperity is really devaluation in
disguise.
Meanwhile, most investors no longer pick stocks; algorithms do.
Trillions flow automatically into ETFs that chase price momentum, not
profit. When liquidity dries up, everything will fall together.
Even the gains AI firms are making come at a cost elsewhere. As the industry sucks up all the available capital, it’s also consuming massive amounts of electricity. Hyperscale data centers use roughly 4.5% of America’s electricity, a figure that could triple in the next two years. That’s raising electricity prices – up about 13% on average between October 2022 and June 2025.
AI centers require significant power to operate and a lot of water to keep them cool, straining local grids and raising utility water prices in areas where these data centers are concentrated. That puts upward pressure on inflation, making both households and businesses more costly to run.
And the supposed AI “revolution that is going to kick off the fourth industrial revolution and even pose an existential threat to humanity? A July MIT study found 95% of AI projects deliver zero profit. Yet trillions in capital keep flooding in. Data-center investment alone could exceed $5 trillion by 2030 — which would require AI services to generate revenues equal to 10% of total U.S. GDP just to break even. Pure fantasy economics and yet the oligarchic insanity over an AI takeover of everything goes on unabated.
On top of all that it emerged last week that AI large language models can quite easily be 'poisoned' and caused to spew out gibberish by hackers.
AI is a massive mis-allocation of capital running in trillions of dollars. Now the data-center theme seems to be going out, people rejecting data centers across the west, due to their energy issues, and effect on fresh water sources.
So many sub-sectors of stock markets have already been falling 6 months or more (transportation, chemicals, materials etc.) but this has largely been concealed because a few big tech corporations heavily invested in AI have propped up the indexes so the headline figures have looked good.
As Wall Street, Lndon and other stock exchanges brace themselves for tech carnage, analysts are starting to wonder if a bigger financial collapse than the dot-com collapse is just around the corner.
AI may be approaching its moment of truth. A technology built to unleash the economy is starting to weigh it down.
The investment flows that have and are still inflating the AI bubble are dependent on monetary flows, and right now we seem to be seeing another liquidity crunch, which will take down everything. The key issue is who is going to save the banks that are too big to fail this time.
