OpenAI Will Burn $115B Before Profitability: Inside the Trillion-Dollar IPO

Numbers That Defy Financial Gravity

My verdict is clear: we're looking at the most ambitious—and potentially dangerous—IPOs in tech history.

OpenAI is targeting a public listing at $1 trillion valuation. Anthropic, its main competitor, just closed a round valuing it at $350 billion. Combined, that's $1.35 trillion in total valuation.

For perspective: that's more than Spain's GDP. More than Apple at IPO. More than Saudi Aramco, which holds the record for the largest IPO in history at $25.6 billion raised.

After X months of hands-on tracking this sector, I won't sugarcoat it: the numbers are extraordinary, but also extraordinarily risky. And as always in finance, the devil's in the details.

OpenAI: The Road to a Trillion Dollars

The Numbers That Matter

If you ask me directly: OpenAI is asking the market to value a company that loses $14 billion annually at $1 trillion. That's a 33x multiple on projected revenue and a negative multiple on profits because, well, there are no profits.

The Restructuring That Changed Everything

In October 2025, OpenAI completed its transformation from nonprofit to Public Benefit Corporation (PBC). The original foundation retains control through the board and received a stake valued at $130 billion.

This was necessary for the IPO. Institutional investors can't buy shares in a nonprofit. Microsoft, which owns 27% of OpenAI after the restructuring (valued at ~$135 billion), was a key driver of this change.

What Sam Altman Really Thinks

In a recent Big Technology Podcast interview, Altman was unusually candid:

"Am I excited to be a public company CEO? 0%... Am I excited for OpenAI to be a public company? In some ways, I am, and in some ways I think it'd be really annoying."

That statement says more than any S-1 document. Altman knows public markets bring scrutiny, quarterly pressure, and shareholders who don't share his long-term AGI vision.

The Investors Betting Big

SoftBank deserves special attention. Masayoshi Son is betting $33 billion on OpenAI after the WeWork disaster. The historical echoes are concerning.

Anthropic: The Faster-Growing Alternative

Numbers Defying Expectations

Anthropic doubled its valuation in 4 months. That's not normal in any market. But the most revealing figure is the profitability projection: 2028, a year or two before OpenAI.

The Second-Mover Advantage

Dario Amodei, Anthropic's CEO and former VP of Research at OpenAI, has built a more efficient company. With half OpenAI's revenue, he projects profitability first.

In Amodei's own words:

"Our revenues grew 10x in the last three years: from 0 to 100 million in 2023, from 100 million to a billion in 2024, and from a billion to 10 billion in 2025. I don't know if that curve will literally continue. It would be crazy if it did."

That last sentence is key. Amodei is more cautious than Altman in his public projections.

The Giants Backing Anthropic

Notably: Amazon, Google, and Microsoft—direct competitors—all invest in Anthropic. This reflects both the potential and the fear of missing out.

Sequoia did something even more unusual: invested in Anthropic despite having a stake in OpenAI. Breaking that venture capital taboo speaks volumes about the opportunity.

The Math That Doesn't Add Up

OpenAI Spends $1.69 for Every $1 It Makes

This is the most important number nobody's talking about. OpenAI has a unit economics problem:

• 2026 revenue: ~$30 billion
• 2026 expenses: ~$44 billion
• Losses: $14 billion

To reach profitability by 2029-2030, OpenAI needs to burn $115 billion in cash. That's more than most Fortune 500 companies are worth.

Anthropic Is 14 Times More Efficient

A Gizmodo analysis puts it brutally:

"OpenAI will burn about 14 times as much money as Anthropic on its way to profitability."

If you ask me directly: this should be reflected in valuations. But OpenAI is worth nearly 3x more than Anthropic. The market is paying a premium for brand recognition, not fundamentals.

Warning Signs the Bulls Ignore

1. Unprecedented Market Concentration

30% of the S&P 500 and 20% of the MSCI World is concentrated in just 5 companies. Adding OpenAI and Anthropic at trillion-dollar valuations would increase that concentration further.

2. Case-Shiller Ratio Exceeds 40

The Case-Shiller P/E stock valuation indicator exceeded 40 for the first time since the dot-com crash. The last time this happened was 2000, right before the collapse.

3. 95% of GenAI Investments Generate Zero Returns

An MIT Media Lab report found that 95% of organizations see zero return on their generative AI investments. If companies buying OpenAI and Anthropic products don't see ROI, how long will they keep paying?

4. AI Represents Half of GDP Growth

AI investments are representing approximately 50% of recent economic growth. If the sector reverses, the macroeconomic impact would be significant.

5. Dell Already Cutting AI Marketing

Dell is reducing its AI-related marketing due to "consumer fatigue." When pickaxe sellers start having problems, it's a sign the gold rush is cooling.

Comparison With Historical IPOs

OpenAI would be 10 times more valuable than Facebook at IPO. Google, which transformed the internet, went public at $23 billion. Has OpenAI transformed the world 43 times more than Google?

The Infrastructure Commitments That Terrify

$1.4 Trillion in Compute Commitments

OpenAI has signed compute infrastructure contracts for $1.4 trillion over 8 years. This includes:

• Data centers with Stargate ($100B+ with SoftBank)
• Cloud capacity with Microsoft Azure
• Nvidia hardware

These commitments are semi-fixed. If revenue doesn't materialize, OpenAI will be trapped paying for infrastructure it can't use.

Data Center Debt Could Exceed $1 Trillion by 2028

According to sector analysis, bank exposure to AI data center debt could exceed one trillion dollars by 2028. That creates systemic risk if the sector cools.

What This Means for Retail Investors

If You Want OpenAI IPO Exposure

You won't be able to buy shares directly until the listing (probably 2027). But you can get indirect exposure:

1. Microsoft (MSFT): 27% of OpenAI
2. Nvidia (NVDA): Critical hardware provider
3. SoftBank (SFTBY): 11% stake, most committed

If You Want Anthropic Exposure

1. Amazon (AMZN): Primary investor and cloud partner
2. Alphabet/Google (GOOGL): 14% stake
3. Salesforce (CRM): Minor strategic investment

My Direct Advice

I won't sugarcoat it: these valuations assume everything goes perfectly. Tech history is full of "too big to fail" companies that failed. WeWork, Theranos, FTX.

If you invest, do it with capital you can afford to lose. The risk/reward asymmetry favors those who wait to see the first earnings reports as a public company.

What Dario Amodei Says About the Future of Work

In a recent statement that flew under the radar, Amodei warned:

"50% of white-collar jobs could disappear in 1-5 years."

He also said "the world is now considerably closer to real danger from AI" than during the peak of safety debates in 2023.

If Anthropic's CEO believes this, why is he building the company that would cause it? Because he believes it's better for Anthropic to lead AI development "safely" than leave the field to less responsible actors.

It's logic you can accept or reject, but at least it's internally coherent.

The Timeline That Matters

Questions Nobody's Answering

1. What If Revenue Doesn't Double Every Year?

Valuations assume indefinite exponential growth. OpenAI went from $6B to $20B in 2025, 233% growth. If that growth moderates to 50% annually, the projections collapse.

2. When Will Competition Start Eroding Margins?

Meta's Llama is open source. Google has Gemini. Amazon has its own models. Competition is increasing while API prices fall. Can OpenAI and Anthropic maintain pricing power?

3. What About Regulation?

Europe has already implemented the EU AI Act. China restricts certain developments. The U.S. is debating legislation. Strict regulation could limit markets or increase compliance costs.

4. What If AGI Arrives Sooner Than Expected?

Paradoxically, if OpenAI or Anthropic achieve AGI (Artificial General Intelligence), the implications are unpredictable. It could be worth $10 trillion or be nationalized for security. Nobody knows.

My Final Analysis

After months of hands-on tracking this sector, my verdict is clear:

OpenAI is overvalued compared to Anthropic using traditional financial metrics. It spends more, loses more, will take longer to be profitable. But it has brand recognition and ChatGPT.

Anthropic is the smarter bet for those who believe in the sector but want a better risk/reward ratio. Less hype, better fundamentals.

Both could fail if the AI bubble bursts before they reach profitability. 95% of companies investing in GenAI see no return. That's not sustainable.

The best strategy for retail investors is indirect exposure via Microsoft, Amazon, or Nvidia. You get upside if the IPOs go well, but you also have diversified businesses that will survive if they don't.

What I wouldn't do: invest my life savings in the biggest and most expensive IPOs in history based on promises that "this time is different." It never is.

---

Data updated January 2026. Sources: CNBC, Bloomberg, TechCrunch, SEC documents, Sam Altman and Dario Amodei interviews.