How much of the stock market's performance is due to AI?

I'm really skeptical of "Business Insider" because they seem to write articles to make stocks move the way they want.

Yeah, we're all totally shocked, right?

But sometimes...

Most college-educated people should have heard about a stock market strategy called "short sales," or just "shorting the market."  "Shorting" means you borrow shares, sell them at the current market price, and hope the stock drops, so when it's time to close the contract you can buy them back at a lower price to replace the shares you borrowed.  So it's a bet that the price of a stock will drop.

Business Insider says an analysis by a company called "S3 Partners" says "short" bets in the S&P 500 are the highest since the company started keeping track of 'em in 2010.  

They claim the value of "short" contracts is almost double what it was a year ago, and virtually the same as it was right before the mortgage meltdown/real estate bust of 2008.

Again, I'm very skeptical of Business Insider because I think many of their claims are designed to manipulate the market.  And "S3 Partners" is a firm whose entire business is selling data to investors.  So, self-serving story.  But even shills can be right sometimes, so here's their theory:

Every day the "newz" is full of stories about more huge investments in AI, and how "the U.S. is in a race with China that we can't afford to lose," eh?  With hype like that, investors have been pouring money into AI stocks, and those companies have been spending like mad--staggering sums.

SO...with the "skirmish" in Iran cutting world oil supplies, and analysts predicting the GOP will get clobbered in November, most people would expect the stock market would be tanking.  Instead it's been doing really well.  

BI claims 60% of the markets' performance is due to AI "and AI-adjacent companies."

If true (and it's likely pretty close), that could be a big problem, because AI has now run into several HUGE problems that threaten to end the party:  First, the big AI companies got user companies hooked with low initial pricing per use.  But because the developers are desperate for revenue, those low-price "hook" offers have ended.  

The result is that big users of AI have seen their monthly bills go from a few thousand dollars a month to a million or so.  So many using companies are sharply limiting use--cutting revenue for AI providers.

Second is that China is offering cheaper versions that are competitive with the latest U.S. AI offerings.  As you could predict, that drives tens of thousands of corporate AI users to the lower-cost Chinese version.

Third is oversupply here at home: Over 1,500 new "data centers" are in various stages of construction or planning in the U.S.  As those come on-line, developers and investors believe they'll be flush with customers and will make a fortune.

By contrast, skeptics believe that when supply exceeds demand, many of these centers will have to compete for customers--by lowering what they charge per token, eh?  

If the second case is true, what do ya think the result will be, eh?  Sure: revenue will be way less than what they assumed.  If that's true, companies that built these massive centers will default on the loans--mean lenders take a huge hit.  What does that do to the market? 

Competitors will buy surplus data centers for a fraction of what they cost to build--but that won't decrease supply.

Fourth is leftist "activists" pressuring corrupt local pols to cancel or re-negotiate contracts for things like water use.  They already signed the contract, but if you haven't noticed, corrupt leftist politicians are happy to tear up contracts if they don't like 'em.

What does that mean?  The new "deal" allowed by the pols will increase costs above what was assumed, making it even harder to make a profit.

So while AI software developers will continue to have significant revenue (though less than expected due to competition from Chyna), companies in the data-center business may find it impossible to stay in business.

SO...four likely "adverse effects," eh?  Even if only one or two of those turn out to be true, what effect do ya think this will have on the market?  And when? 

Huge systems have equally huge inertia, meaning something can be "fatally shot," but because of the huge inertia it takes months for the effects to be felt by the whole economy.  If the problems discussed above are accurate, the damage won't be felt right away--but will continue to have what the Brits call "knock-on effects" for a few years.  

Of course I could easily be wrong.  And again, I'm very skeptical of BI and "S3 Partners." 

Source.

Source 2.