You’ve been told AI is the next gold rush. That OpenAI and Anthropic are the new Microsofts. That missing this wave is like missing the internet.
But here’s the uncomfortable truth nobody wants to admit: the companies leading the charge are built on hype, not profits. And that’s a problem when the bill comes due.
Every week, another headline screams about a $100 billion valuation. Every month, another round of funding from investors who are terrified of being left behind. But ask yourself: when was the last time you saw a financial statement from OpenAI? When did Anthropic ever show a path to profitability that wasn’t just ‘more compute, more users, more magic’?
I’ve been following this space for years. I’ve talked to founders, VCs, and analysts who all whisper the same thing behind closed doors: these companies are too big to ignore, yet too unproven to trust. And that contradiction is about to explode.
Let’s start with the numbers. OpenAI is reportedly worth $80–100 billion. That’s more than most Fortune 500 companies. But according to leaked documents, it’s burning through cash at a rate that would make a startup blush. Revenue is growing, sure — but it’s a fraction of the cost. Every API call, every ChatGPT session, every server rack is a loss leader. The real money? Nobody knows where it’s coming from.
Anthropic is even more opaque. They’ve raised billions from Google, Salesforce, and others. But what’s the moat? A slightly safer model? A nicer chatbot? In a world where Meta’s Llama is free and open-source, how do you justify a $30 billion valuation?
This is the paradox of the AI industry: the technology is revolutionary, but the business models are pre-revolutionary. These companies are trying to sell shovels in a gold rush where the gold hasn’t been found yet.
Now, the IPO question. The market is hungry for the next big tech float. But the rules have changed. Post-2021, investors want profitability, not just growth. They want moats, not just hype. They want to see a path to $10 billion in revenue, not just a slide deck with arrows pointing up.
OpenAI and Anthropic face a brutal choice: go public now and risk a catastrophic valuation haircut, or stay private longer and hope the business model catches up. Neither option is good.
Here’s the twist most people miss: the real bottleneck isn’t technology — it’s trust. Public markets will punish these companies for lacking the sustainable moats that justify their private prices. The SEC will ask hard questions about revenue recognition, regulatory risks, and the fact that a single government policy could gut their entire business model.
Remember when WeWork tried to IPO? The hype was real, the valuations were insane, and then the public market said, ‘Show me the money.’ The same thing is coming for AI’s darlings.
So what’s the play? For investors, don’t be a bag holder. For founders, build a real business, not just a model. For everyone else, buckle up: the AI bubble won’t burst with a crash — it will deflate slowly, as investors realize that building a moat is harder than building a model.
And that’s the truth nobody wants to put in a press release.
FAQ
Q: Aren't OpenAI and Anthropic generating massive revenue already?
A: Not really. Revenue is growing from subscriptions and API usage, but it's a tiny fraction of the massive costs — compute, talent, data, and R&D. Both are likely burning billions annually with no clear path to profitability in the near term.
Q: What's the practical implication for investors?
A: If you're holding shares in secondary markets or betting on an IPO pop, expect disappointment. These companies will either have to accept much lower public valuations or delay IPOs for years, tying up your capital with no liquidity event in sight.
Q: Couldn't these companies pivot to a profitable business model?
A: Theoretically, yes. But pivoting means slowing down the arms race with competitors offering free or cheaper models. Any move to monetize aggressively could alienate users and developers, while the open-source alternatives keep getting better. The window for a clean pivot is closing fast.