Three years after igniting the AI revolution, OpenAI is "forced" to go public.

Three years after igniting the AI revolution, OpenAI is "forced" to go public.

```

In November 2022, ChatGPT launched and its servers were overwhelmed on the first day. Before this, no technology company had ever acquired 100 million users at such an astonishing speed.

From that moment, the “AI revolution” went from being an industry term to a reality that ordinary people could feel.

The curtain raiser for this revolution was OpenAI.

Three years later today, according to The Information and other media reports, OpenAI is preparing to secretly submit its IPO draft to the SEC, aiming to go public as early as September, with a target valuation of over $1 trillion. Altman has not denied this report.

But just 48 hours ago, he told employees in an all-hands meeting: Submitting an IPO application and truly being ready for listing are two different things. The company will not rush into the public market before conditions are mature.

These two sentences make OpenAI's current situation very clear.

Putting Ducks on the Rack

Altman says they're not ready, but reportedly the company is submitting forms today. To understand this contradiction, you must look at the three layers of pressure behind it.

First, Anthropic is overtaking.

In the past 15 months, Anthropic’s annualized revenue grew from $1 billion to $30 billion, a 30-fold increase. In the same period, OpenAI's revenue grew from about $20 billion to $25 billion, up roughly 25%.

In April this year, Anthropic’s annualized revenue officially surpassed OpenAI’s.

The more embarrassing numbers are in Q2. Anthropic expects about $10.9 billion in Q2 revenue, and about $600 million in operating profit, turning profitable ahead of OpenAI.

Anthropic’s private valuation is already $1.2 trillion, higher than OpenAI’s self-imposed $1 trillion IPO target.

Anthropic is also preparing to go public, as early as October. Altman has privately expressed hope that OpenAI will make it to the public market first.

Second, SpaceX is about to grab the money next month.

SpaceX, owned by Musk, plans to raise about $75 billion at a $1.75 trillion valuation next month—this will be the largest IPO in history.

According to reports citing insiders, there is “a lot of game” between OpenAI and SpaceX. OpenAI hopes investors will reserve funds for itself in advance and not bet all their firepower on SpaceX.

The competition for public market funds between the two companies is real.

Musk was previously a co-founder of OpenAI, later left and founded the competitor xAI, and the two companies are currently engaged in a lawsuit.

Now, Musk’s other company wants to grab water from the same pool.

Third, computing power is a bottomless pit, and money is not enough.

OpenAI expects to spend $665 billion on computing power by 2030. The company has so far raised nearly $200 billion, but is still far from that figure.

IPO fundraising is the next window for replenishing ammunition, with no alternative.

With these three pressures combined, the logic is clear: OpenAI isn’t ready, but can’t afford to wait.

Internal Disunity

This IPO has not achieved a unified pace internally within the company.

Major investors are surprised by the speed, which is not a signal that should appear before a normal IPO.

OpenAI CFO Sarah Friar’s attitude towards the pace of listing is noticeably more cautious than Altman’s, which has been specially reported, indicating genuine internal differences.

Altman himself mentioned in the all-hands meeting that SpaceX’s upcoming large-scale listing and global economic trends will be key external variables affecting whether OpenAI can ultimately go public. In other words, he himself is not sure if the September window will open.

This does not look like a company confidently charging toward IPO. This is more like: The situation is forcing them forward, but even internally, nobody can say how far they’ll get.

Cracks in the Story

The deeper problem is that OpenAI’s story for the public market is becoming increasingly hard to stand up.

ChatGPT’s initial explosive fame established OpenAI’s unshakable brand advantage on the consumer side. But in the past 12 months, ChatGPT's web traffic market share has fallen from 87% to 68%, while Google Gemini has increased from 5% to over 18%. Its monthly active users have grown from 350 million to 750 million.

Google has search, Android, and Chrome—these are entry points OpenAI doesn’t have and can’t buy.

On the enterprise side, OpenAI is also under pressure. In direct head-to-head competition, Claude has won roughly 70% of enterprise orders.

Anthropic, with fewer users (134 million vs OpenAI’s ~900 million), accounted for a higher market revenue share (31.4% vs 29%), with a monthly average revenue of $16.2 per active user—commercial efficiency is ahead.

Both directions of lost ground happened in OpenAI’s core base.

OpenAI’s response is to bet on the AI Agent economy. If AI Agents become the main mode of interaction between humans and machines, OpenAI’s model will be the operating system of this economy, and $280 billion in 2030 revenue will be only a small part of the ecosystem.

This logic stands up conceptually. But it requires OpenAI to maintain growth on all three fronts—consumer, enterprise, and developer ecosystem—while rivals on all three fronts are accelerating, and OpenAI’s own growth rate is only 25%.

To go from $25 billion to $280 billion in 4 years—a 65% annual compounded growth rate—these are the numbers the public market will scrutinize.

The Most Expensive Mirror

This is the first time OpenAI must open its books.

Previously, the $852 billion private valuation was built on the shared narrative of a few institutional investors, with opaque financial data and inconsistent valuation methods. Once the S-1 is public, loss amounts, cost structure, Microsoft dependency clauses, and the specific assumptions for the $280 billion forecast will all be laid bare to the public market.

The public market’s questions are straightforward: Can OpenAI win, when can it turn a profit, and why is it worth this price?

This is exactly what Altman means by “conditions aren’t mature.” He knows better than anyone that the private market buys stories, the public market buys numbers. And OpenAI’s current numbers—slowing revenue growth, shrinking market share, still losing money—are not the best entrance conditions.

But it has no way to wait for the best entrance conditions.

With every quarter that Anthropic goes by, the gap closes. With every day SpaceX moves forward, there’s less money in the market. The computing power bill is increasing every month.

This is what “riding a tiger” means: it’s easy to get on, hard to get off.

OpenAI has made itself the biggest symbol of this round of AI revolution in three years, and now that symbol itself has become a burden—it must live up to $1 trillion, even if it’s not ready.

Asset Impact

Microsoft is the biggest beneficiary of this IPO. With an investment of $13.8 billion corresponding to 27% equity, if the IPO lands with a $1 trillion valuation, the paper gain exceeds $270 billion. Short-term positive, buy logic is clear.

But note: After listing, OpenAI will seek to reduce reliance on Microsoft, and Microsoft’s exclusive AI advantage faces a dilution risk in the medium to long term.

Nvidia: OpenAI is one of the largest buyers of computing power. IPO fundraising → acceleration of computing power spending → support for GPU demand. But in the medium to long term, OpenAI has an in-house chip roadmap—this is a discount variable in valuation.

CoreWeave: OpenAI’s computing power partner, directly benefiting from accelerated computing power spending, can be used as a proxy target.

A-share AI sector: If the $1 trillion pricing succeeds, it will anchor global AI valuations positively; if the IPO is discounted or delayed, the AI bubble narrative will face temporary pressure.

Follow-up Tracking

  • SpaceX IPO (next month): Will it drain funds from the public market, will OpenAI roadshow be forced to adjust timetable.
  • Anthropic’s Q2 results: If $10.9 billion revenue + profit is fulfilled, it will form direct narrative suppression before OpenAI’s roadshow.
  • Anthropic IPO timing (earliest October): Whoever goes public first, whoever prices higher, will be this year’s most important pricing event for the AI sector.
  • S-1 public version: Loss amount, Microsoft contract terms, breakdown of $280 billion forecast assumptions—any number beyond expectations could trigger market re-pricing.
  • Whether Altman changes his timing judgment: He said market conditions are the ultimate variable.

Altman brought ChatGPT to the world and started this AI revolution. But the revolution will not wait for one person to be ready.

Now, he sits on a tiger that’s already running, with Anthropic, SpaceX, and Google ahead, and a bottomless pit of computing power bills following behind.

Getting off is hard. Not getting off is also hard.

Risk Disclaimer and Exemption ClauseThe market carries risks, and investments should be made with caution. This article does not constitute personal investment advice and does not take into account the individual investment objectives, financial situation, or needs of any user. Users should consider whether any opinions, views, or conclusions in this article are suitable for their particular circumstances. Investing based on this is at your own risk. ```