Skip to content

Instantly share code, notes, and snippets.

@johnnymo87
Last active December 10, 2025 17:03
Show Gist options
  • Select an option

  • Save johnnymo87/2dd7fb5ec3f078eff0dbfa15197905c9 to your computer and use it in GitHub Desktop.

Select an option

Save johnnymo87/2dd7fb5ec3f078eff0dbfa15197905c9 to your computer and use it in GitHub Desktop.
A summary of Zvi Mowshowitz's analysis on OpenAI's financial situation (Nov-Dec 2025)

OpenAI's Financial Situation

Your brother mentioned hearing OpenAI isn't turning a profit. Here's what Zvi has written:

OpenAI's Losses and Cash Burn

From AI #133: America Could Use More Energy (Sep 11, 2025):

"OpenAI is now projecting that it will burn $115 billion (!) on cash between now and 2029, about $80 billion higher than previously expected."

From AI #128 (Aug 7, 2025):

"OpenAI took a $5 billion loss in 2024, but they are tripling their revenue from $4 billion to $12 billion in 2025. If they (foolishly) held investment constant (which they won't do) this would make them profitable in 2026."

So yes - OpenAI is massively unprofitable, but that's because they're investing aggressively in compute and research, not because the business model doesn't work. Their revenue is growing rapidly ($4B → $12B → projected $20B annualized).

OpenAI's Financing Strategy (Not a Google Buyout)

Rather than being acquired, OpenAI has been raising massive amounts of capital. From OpenAI Shows Us The Money (Sep 24, 2025):

  • Nvidia investing up to $100 billion in OpenAI
  • Stargate project: $500 billion, 10-gigawatt commitment with Oracle and SoftBank
  • $1 trillion buildout of computing warehouses planned

"This is another big loss for Microsoft. Microsoft could've been OpenAI's capital partner of choice as they got in early and had lock-in. But Microsoft wasn't willing to keep delivering and backed out."

The Corporate Restructuring Drama

OpenAI has been converting from a nonprofit to a for-profit structure. From OpenAI Moves To Complete Potentially The Largest Theft In Human History (Oct 31, 2025):

"OpenAI is now set to become a Public Benefit Corporation, with its investors entitled to uncapped profit shares. Its nonprofit foundation will retain some measure of control and a 26% financial stake, in sharp contrast to its previous stronger control and much, much larger effective financial stake. The value transfer is in the hundreds of billions..."

OpenAI is valued at $500 billion and planning an IPO at potentially $1 trillion.

What About a Google Buyout?

Zvi hasn't discussed a Google acquisition of OpenAI, and it seems highly unlikely because:

  1. Google is a direct competitor with Gemini - they're racing against OpenAI, not looking to acquire them
  2. Antitrust concerns would make such a merger essentially impossible
  3. OpenAI has plenty of funding from Nvidia, SoftBank, and others
  4. The Microsoft relationship has soured but they still hold ~27% of OpenAI

The more realistic concern Zvi raises is about the Nvidia-OpenAI relationship creating problematic incentives:

"A key worry has long been that OpenAI depends on Nvidia, so OpenAI may be reluctant to speak about AI risk lest they anger Nvidia and endanger their chip allocations."


Bottom Line

The Tom's Guide style head-to-head comparisons are roughly correct that Gemini 3 wins on many benchmark tasks, but Zvi's deeper analysis suggests those wins come with significant tradeoffs in reliability and honesty. The "best" model depends heavily on:

  1. What you're doing (coding vs research vs creative work)
  2. How much you need to trust the output (Gemini hallucinates more)
  3. Whether you want a collaborator or a tool (Claude has more "soul")

For most general purposes as of December 2025, Zvi recommends Claude Opus 4.5.

As for OpenAI's finances: they're unprofitable by choice - deliberately burning cash to build infrastructure. They've raised enough capital ($100B+ from Nvidia alone) that they're not at risk of needing a bailout or acquisition. A Google buyout is essentially off the table due to antitrust and competitive dynamics.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment