Article image

The emperor's new algorithm runs on venture capital.

I remember standing in a Palo Alto juice bar in 1999 listening to two guys debate whether Pets.com would crack a $10 billion market cap post IPO. They weren't laughing. That memory surfaced this week when news broke about yet another artificial intelligence unicorn charging toward Wall Street with numbers so ludicrous they'd embarrass a Nigerian prince email scam. The champagne corks are popping for Anthropic's alleged $300 billion valuation target, a tidy 63% premium over its last funding round three months ago. Because obviously that's how business fundamentals work in the 2020s.

What fascinates me isn't the astronomical sum, but the audacity required to float it with straight faces. For comparison, Disney's current market cap hovers near $230 billion. The entire global film industry generates about $100 billion annually. But sure, an AI company that most people couldn't name until last Tuesday deserves a valuation surpassing every movie studio, theme park, and streaming service combined. Nothing to see here.

The cognitive dissonance would be hilarious if real money weren't involved. These firms position themselves as ethical alternatives to the robber baron tech giants while racing to billionaire creation speeds that would give Rockefeller whiplash. I've sat through enough startup demos to recognize the pattern, the fever dream pitch decks claiming this time is different while recycling every mistake from the dot com crash. Remember when WeWork was going to revolutionize physical space valuation at $47 billion? Theranos at $9 billion for blood tests that didn't work? Juicero at $700 million for a wifi connected fruit press? The museum of tech delusion keeps adding wings.

Here's what nobody's discussing at the VC pool parties. For all the breathless talk about transformative AI, we haven't seen a single foundational model company achieve profitability. Not one. The math is delightfully absurd a 2025 UC Berkeley study estimated that training Anthropic's Claude 3 model cost more than the GDP of Micronesia. Each query reportedly burns cash faster than a Bitcoin miner with cheap electricity. Yet suddenly investors are scrambling to pay five times revenue for companies with negative margins? The ghost of Pets.com just high fived the specter of Webvan.

Meanwhile, Wall Street's enablement apparatus snaps into action. Wilson Sonsini didn't earn their $1,500 hourly rates by asking awkward questions about unit economics. The white shoe law firms and bulge bracket banks have refined the IPO industrial complex into a carefully choreographed ballet where due diligence means checking if the CEO's yacht is photo ready for the prospectus cover.

Let's address the human toll hiding behind the glowing headlines. Last month I had coffee with an Anthropic engineer who looked like he hadn't slept since the Clinton administration. When I asked about IPO rumors, he stared into his oat milk latte like it held existential answers. The technology breakthroughs driving this gold rush are being built on the backs of brilliant, burned out twenty somethings coding through panic attacks. They're not getting $300 billion paydays. Those golden tickets flow exclusively to the Sand Hill Road crowd who've perfected the art of flipping private shares like day traders with better suits.

The regulators? Don't make me laugh. The SEC still hasn't fixed the basic SPAC loopholes that allowed a parade of 2021 era frauds to fleece retail investors. Now they're expected to police AI accounting voodoo while lobbyists ensure every proposed safeguard dies in committee. Meanwhile, Elizabeth Warren tweets sternly about accountability before boarding another flight to Martha's Vineyard. Performative outrage is Washington's highest grossing export.

Normal people pay the price when these paper castles collapse. I think about my barber, a sweet Nicaraguan grandmother who invested her life savings in Rivian at $150 based on CNBC hype. It now trades below $20. The next AI crash won't vaporize just VC play money, but teacher pensions, municipal bond funds, and middle class retirement accounts. The devastation will make 2008 look like a minor accounting error.

Historical amnesia remains Wall Street's greatest superpower. Today's bankers weren't around for the 2000 dot com implosion that wiped out $5 trillion in wealth. They learned about it in MBA case studies between tequila shots at Cancun spring break. Now they cheerlead a company whose valuation rose more in three months than IBM did in three decades as if math itself has been disrupted.

The most pathetic part? These firms justify astronomical valuations by positioning themselves as saviors. OpenAI claims it's building God like intelligence to solve climate change while accepting Saudi oil money. Anthropic touts constitutional AI principles while its investors push for exit velocity over ethics. It's like watching Bernie Madoff rebrand as a spiritual guru.

Make no mistake. I believe artificial intelligence will transform society more profoundly than steam engines or smartphones. But this IPO frenzy has nothing to do with technological progress and everything to do with financial engineering. Real innovation happens incrementally, through peer reviewed papers and open source collaboration, not VC beauty contests judged by spreadsheet jockeys who think LLaMA is an animal at the zoo.

The next twelve months will either validate this madness or expose it as history's greatest Ponzi scheme dressed in machine learning jargon. Personally, I'm stocking up on popcorn and preparing my best told you so smile. The juice bar debates haven't changed in twenty five years. Only the buzzwords have.

Disclaimer: The views expressed in this article are those of the author and are provided for commentary and discussion purposes only. All statements are based on publicly available information at the time of writing and should not be interpreted as factual claims. This content is not intended as financial or investment advice. Readers should consult a licensed professional before making business decisions.

Daniel HartBy Daniel Hart