Silicon Valley's two brightest AI stars, OpenAI and Anthropic, have both run into a wall on Wall Street.
OpenAI sought to raise $30 billion, but investors were only willing to commit $11 billion — less than half. The company had no choice but to cut its fundraising target by one-third. Meanwhile, Anthropic, on the verge of what could be the largest IPO in history, was stabbed in the back by one of its major investors: "Bro, you might get slapped with a 'junk' rating."
But that's not the end of it. This chill has already spread from the primary market to the secondary market at lightning speed — the Nasdaq has crashed.
On Friday's close, the Nasdaq fell 3.46%, marking its largest single-day drop in months. NVIDIA (NVDA) and Advanced Micro Devices (AMD) dragged the entire chip sector down with them. Amazon.com (AMZN) saw its cash flow completely burned through by AI infrastructure spending. Intel (INTC), reeling from a disastrous earnings report, plummeted nearly 30% in a single day.
A full-blown winter has set in. The Nasdaq fell about 5.3% for the entire month of October, its worst month since March 2025. Everything looks exactly like the eve of the internet bubble burst 26 years ago.
Even Altman Can't Borrow Anymore
Hard to believe. OpenAI — the OpenAI of GPT-4 and Sora, the OpenAI led by Sam Altman — actually stumbled in fundraising?
According to the latest insider revelations from The Wall Street Journal, Altman was as confident as ever this time, with a new funding round targeting a whopping $30 billion. The result — only $11 billion raised so far. Less than half the target. The company's valuation has also shrunk to $1.25 trillion. Incredible.
Keep in mind, this is OpenAI. Over the past year, every VC and every hot money fund was fighting tooth and nail to get in. But now, investors have suddenly become extremely picky, even nervous. Everyone is worried about just how much money AI will burn and whether it will ever pay off. Even someone as powerful as Sam Altman, a fundraising superman, now has to lower his posture, saying he will continue working hard over the coming weeks to "solicit donations" from family offices and high-net-worth individuals.
Something is off. Very off. If OpenAI's setback were just an isolated case, then the plight of another AI giant would truly send chills down your spine.
Anthropic Is Done For Too
As everyone knows, Anthropic, OpenAI's most hated competitor, is about to IPO — and it could be the largest IPO in human history. For this IPO, they've planned an extraordinarily ambitious spending blueprint: from now until 2045, they plan to spend over $800 billion on computing infrastructure.
Where does all that money come from? Borrowing, of course. To borrow such a massive amount at low cost, the company must have a good credit rating — the technical term is "Investment-Grade." It's like your personal credit score: with a high rating, banks are willing to lend to you at very low interest rates.
But just at this critical moment, something went wrong. One of their major investors, private equity giant Blackstone, suddenly changed its tune. Blackstone had been telling other investors, "Don't worry, Anthropic will definitely secure an 'Investment-Grade' rating before the end of the year." Now they're telling investors that, based on signals from major rating agencies, Anthropic will likely fail to obtain an "Investment-Grade" rating after going public.
So what will it get instead? "High Junk" — high-yield junk grade. That's something else. What does "junk" mean? It means that in the eyes of professional institutions, your default risk is high. Many pension funds and insurance funds are simply not allowed by regulation to touch this type of "junk bond." The rest, those willing to lend to you, will inevitably demand much higher interest rates.
Think about it: originally planning to borrow $800 billion, with interest rates jumping from 3% to 6% — what does that even mean? The company's entire financial model would simply collapse.
Many plans would have to be scrapped and redone due to budget constraints. I came across an analyst's post that put it perfectly: companies that missed targets, fell short of expectations, or failed to execute are being punished. "Punished" — that word is so precise.
Whether it's OpenAI's fundraising setback or Anthropic's rating dilemma, it's essentially the capital market repricing risk. People are starting to get scared. Scared of this endless "cash burn," scared that the money they've invested will ultimately buy nothing but a pile of expensive servers and a growing line of loss figures.
The Avalanche Begins
You thought that was it? No. On the very day the Financial Times broke the news about Anthropic's rating problem, panic over the AI bubble spread like a virus through the secondary market — the Nasdaq 100 index plunged 3.5% at close, its largest single-day drop since June.
Wall Street is on edge. What does that mean? To use one fund manager's words: it's a complete bloodbath.
Why is this happening? At the end of the day, the root cause of everything is still AI. This AI race is essentially a crazy money-burning game. Buying chips, building data centers — none of it comes cheap. But previously, everyone assumed that as long as the story was sexy enough, endless cheap money would flow in from the capital markets.
But now, Wall Street is telling you with actions: even the landlord's family has no surplus grain left. To better understand this drama, let's rewind the timeline a bit.
Actually, before the Anthropic news came out, market sentiment was already terrible. The main reason: everyone suddenly realized that the big backers of AI companies don't seem to be doing so well either.
Let me give you an example: Amazon.com (AMZN). On the surface, its AWS cloud business saw revenue surge 39% — incredibly strong. But on closer look, something doesn't add up. Amazon's capital expenditure on AI data centers and related infrastructure this year, driven by frenzied spending on AI servers, is projected to reach $220 billion. Amazon, which had $14.8 billion in positive cash flow a year ago, now has a latest trailing twelve-month free cash flow of -$42.4 billion.
Microsoft, Meta, Google... these so-called "hyperscalers" have propped up over one-third of S&P 500 capital expenditure over the past year. They are the main engines of this AI boom. And now you're telling us the giants are out of money too?
In the past, people thought these tech giants had plenty of cash and could single-handedly sustain the AI arms race. But now, as they increasingly rely on external financing, and as even a star company like Anthropic faces a credit crisis, everyone is starting to worry. Can this game continue? Is there enough money to keep burning?
So the selling began. Microsoft, Meta, Google... all fell together. Intel (INTC), its stock price in freefall, plunged 28.6%. Even Reddit dropped 13%.
It's quite emotional, folks. A groundbreaking technology emerges, everyone goes crazy for it, believing it will change the world. Countless companies, basking in the glow of their labs, paint enormous pies, and even with zero profit, their valuations soar to the heavens. Data centers and everything else are built at breakneck speed.
Until one day, everyone suddenly realizes that these companies are burning money far faster than they can earn it. I've always believed that AI is the future. But the road to the future has never been a straight line. It is inevitably filled with bubbles, hype, and painful shakeouts.
The wall that OpenAI and Anthropic hit on Wall Street today may be the beginning of that shakeout. The era of demanding money from investors like an entitled lord just because you have "AI" in your name — is over.
One More Thing
Honestly, sometimes I feel like history really does repeat itself. Everything happening right now — I can't shake the feeling I've seen it somewhere before. Same smell, same recipe, all too familiar... Oh, right. I remember now. 26 years ago, in 2000 — the internet bubble.
Comments