When someone tells you an $830 billion valuation might actually be too low, you're entitled to some skepticism. But that's exactly what Deepwater Asset Management's Gene Munster argued on Friday about OpenAI, despite the fact that the company is projected to burn through somewhere between $100 billion and $150 billion in losses over the next few years.

The math here feels absurd at first glance. OpenAI's valuation has rocketed from roughly $30 billion in 2022 to a potential $830 billion today—that's about 28 times higher in just four years. And yet Munster thinks the market is still underestimating what AI could become.

When Losses Don't Tell The Whole Story

Writing on X, Munster called it "the mind-numbing reality of AI's potential" and said OpenAI remains "undervalued" even at these eye-watering numbers. According to a Wall Street Journal report he cited, OpenAI is seeking to raise about $100 billion at an $830 billion post-money valuation by the end of March 2026.

Sure, there's a bear case here. The company is hemorrhaging cash. Internal forecasts suggest cumulative losses could hit $100 billion to $150 billion between 2025 and 2029. There are uncomfortable parallels to the dot-com era, when early internet darlings like Netscape and Lycos dominated headlines before disappearing entirely. And the talent war is heating up, with Meta Platforms Inc. (META) and Alphabet Inc.'s (GOOGL) Google aggressively poaching top AI researchers.

But Munster thinks the bullish case deserves equal weight, and his reasoning comes down to who's placing the biggest bets.

Follow The Money—And The CEOs

When the CEOs of Microsoft Corp. (MSFT), Nvidia Corp. (NVDA), Meta, Alphabet, and Tesla Inc. (TSLA) all start reallocating tens of billions of dollars toward AI, that tells you something. Munster pointed to Satya Nadella, Jensen Huang, Sundar Pichai, Mark Zuckerberg, and Elon Musk as leaders who are essentially "betting the company" on artificial intelligence.

These aren't speculative side projects. When executives of this caliber move that kind of capital, Munster argues, it signals a belief that AI's substance will eventually overtake the hype. And if they're right, OpenAI's current valuation starts to look more reasonable.

The Growth Numbers Are Wild

Here's where things get interesting. Munster estimates OpenAI's revenue could climb from about $4 billion in 2024 to $35 billion in 2026 and $70 billion by 2027. That kind of growth trajectory is rare even among the fastest-growing public tech companies.

Then there's the scale. ChatGPT now has close to 900 million weekly active users, according to Munster's estimates. That's a distribution advantage matched by basically nobody except Google and Meta. When you combine that user base with triple-digit revenue growth, the valuation premium starts to make more sense.

Munster's take boils down to this: if OpenAI can sustain 100% growth rates, $830 billion will look cheap in hindsight. If it can't, the valuation is headed lower. It's binary, but at least it's honest.

A Trillion-Dollar IPO On The Horizon

OpenAI isn't content to stay private forever. Reuters reported that the company could file with U.S. securities regulators as soon as the second half of 2026 for a blockbuster public debut. Estimates suggest the IPO could value OpenAI at up to $1 trillion, which would make it one of the largest market debuts in history.

Meanwhile, OpenAI is reportedly in talks with Amazon.com Inc. (AMZN) over a potential investment of more than $10 billion. The discussions also include exploring the use of Amazon's AI chips, which would deepen the partnership between the two companies and potentially reduce OpenAI's reliance on other chip suppliers.

For context, Microsoft (MSFT) maintains a favorable long-term outlook according to market data, even as its short and medium-term performance trends remain under pressure. That matters because Microsoft is already one of OpenAI's biggest backers, and the success of that bet could significantly impact Microsoft's future growth trajectory.

So Is OpenAI Really Undervalued?

The case for OpenAI being undervalued at $830 billion rests on a few assumptions. First, that the company can continue growing revenue at rates that would make most CFOs dizzy. Second, that its massive user base translates into sustainable competitive advantages. And third, that the collective judgment of Big Tech's leadership is correct about AI's long-term potential.

None of those assumptions are guaranteed. The losses are real. The competition is fierce. And history is littered with companies that dominated emerging technologies before getting overtaken or obsolete.

But if Munster is right—if AI really does live up to the "mind-numbing" potential he describes—then even a valuation that has grown 28-fold in four years might still be leaving money on the table. And that's either the most exciting investment thesis of the decade or a cautionary tale waiting to happen.