$1.2 trillion. That number is roughly 40% above the $852 billion post-money valuation from March's mega-round. OpenAI is pushing its final pre-IPO funding round, targeting a valuation that kisses $1.2T. Everyone who piled into March's $122 billion super-round — Microsoft, Nvidia, SoftBank — is nodding along. They're not agreeing OpenAI is worth $1.2T. They're protecting their March bet: if $1.2T sticks, their $852B entry is already up 40% on paper.

An Investor-Initiated Trillion-Dollar Pricing

Here's the detail worth pausing on: this round wasn't OpenAI's idea. Investors initiated it. March's $122 billion raise was one of the largest private funding rounds in history, with Microsoft (~27% stake as largest single shareholder), Nvidia (committing $300 billion in compute credits), and SoftBank (as anchor investor, having secured a $400 billion bridge loan) deeply embedded. That March round wasn't a simple capital injection — it was a coordinated industry supply-chain play.

Now investors are proactively pushing for a $1.2T valuation in a new round. The logic is straightforward: if you put money in at $852B in March, you need a higher-valuation follow-on to "lock in" paper returns on that earlier investment. If the $1.2T round succeeds, March investors are sitting on ~40% unrealized gains, and the IPO pricing has a higher anchor. This isn't paying for OpenAI's future — it's paying for investors' own portfolio math.

Revenue Growth vs. Loss Expansion: A Race OpenAI Is Losing

To make $1.2T pencil out, OpenAI needs to show a convincing set of financials. Here's where things stand: annualized revenue around $25 billion, with strong growth. Simultaneously, Q1 2026 loss of $9.3 billion, Q2 loss expanding further to $12.3 billion. Combined, that's over $21.6 billion in losses in H1 2026 — meaning for roughly every dollar OpenAI earns, it burns about one dollar.

What makes this uglier: Q2's single-quarter loss of $12.3B is nearly double the quarterly revenue run rate. The burn rate is accelerating, and while revenue is growing, the growth isn't outpacing the losses. For investors, the only "safety net" is the slope of the revenue growth curve — if revenue keeps surging, losses will eventually be diluted. But "eventually" is doing a lot of heavy lifting. Before IPO, investors need to see that curve inflect.

The core metrics propping up the revenue story are ChatGPT's user numbers: 900 million weekly active users, monthly actives breaking past 1 billion. This is one of the fastest-growing consumer AI products in human history. More importantly, enterprise revenue now accounts for over 40% of total revenue — meaning OpenAI isn't just a chatbot app for consumers. It's becoming enterprise AI infrastructure. Enterprise customers pay more willingly, sign longer contracts, and generate more predictable revenue. That 40% enterprise mix is the most critical pillar of the $1.2T narrative.

Competition Reversal Signals

Complicating the $1.2T narrative further are the competitive dynamics.

Anthropic's annualized revenue has hit $30 billion, surpassing OpenAI's $25 billion. On the most basic competitive metric — who makes more money — OpenAI is no longer in first place. For a company eyeing an IPO, being overtaken in revenue by a competitor isn't a signal you can brush off.

More unsettling is the market share shift. ChatGPT's global AI assistant market share has dipped below 50%, settling at 46%. In the AI assistant category, ChatGPT was once the absolute monarch. 46% means more than half of all AI assistant usage is now happening outside ChatGPT — Google's Gemini, Anthropic's Claude, and a swarm of open-source and vertical models are siphoning users.

Revenue surpassed, share declining, but valuation needs to climb 40% — the logic gap demands an explanation. OpenAI's case might be: ChatGPT has 1 billion MAU, an unmatched distribution moat; 40% enterprise revenue proves commercialization is working; expanded losses are normal during the investment phase. But investors must judge for themselves: do these explanations describe a consolidating monopoly or an eroding first-mover advantage?

IPO Pushed to 2027: Time Isn't on OpenAI's Side

OpenAI has explicitly pushed its IPO timeline to 2027. The surface reasoning: room for safety commitments — going public means quarterly earnings pressure, and staying private preserves room for bolder safety investments. But the more pragmatic reason may be that current financials can't support an IPO valuation that satisfies all stakeholders.

If OpenAI IPO'd today, investors would see $25B annualized revenue, $10B+ quarterly losses, and declining market share. What public-market valuation does that combo support? Maybe $800B, maybe $1T, but $1.2T would be a stretch. By pushing to 2027, OpenAI hopes to show: revenue catching up to burn, enterprise share expanding further, ChatGPT's market leadership reconsolidated.

But the window is finite. Anthropic won't wait. Google won't wait. The open-source community definitely won't wait. The competitive landscape in 18 months could be far more brutal than today.

OpenAI remains one of AI's most influential companies. One billion monthly ChatGPT users is an undisputed distribution moat; enterprise revenue growth proves commercialization works. But a $1.2T valuation demands more than "influential" — it demands "irreplaceable." And when a competitor surpasses you in revenue, erodes your share, and your losses accelerate, the word "irreplaceable" gets increasingly expensive.

A $1.2 trillion valuation doesn't price OpenAI's revenue — it prices the belief that the AI buildout won't brake. And unlike revenue, beliefs can change.

See you tomorrow.