OpenAI Files for IPO as Altman Outlines "Phase Three" Strategy
Claire Weston
OpenAI confidentially filed its S-1 on Monday while Altman published a manifesto calling this the company's 'Phase Three' — a pivot from product expansion to universal AGI, framed squarely for the public-market investors it will soon face.
Why file now?
OpenAI submitted a confidential S-1 to the SEC. Goldman Sachs and Morgan Stanley are lead underwriters; the target window is this fall.
The company said it "expected the filing to leak, so decided to announce it" itself — while admitting the timeline is uncertain and that "some things are easier as a private company."
This means → Filing ≠ imminent listing, but it secures the option to move fast. The real catalyst: rival Anthropic filed its own confidential IPO application a week earlier.
What exactly is Altman's 'Phase Three'?
Altman and chief scientist Pachocki co-authored a blog post splitting OpenAI's history into three acts: Phase One built AGI foundations, Phase Two shipped products globally, and Phase Three = "the economy is reshaping itself around AI."
Three explicit goals: build automated AI researchers, accelerate economic growth, and give every person on earth a personal AGI.
In plain terms = This is not a technology roadmap. It is a belief statement written for future shareholders — wrapping commercial expansion inside a social-mission narrative so investors buy the vision.
What did the safety statement say — and not say?
The pair wrote: "Fully automating everything is not the future we want — it is both unsatisfying and dangerous." They called for a global body empowered to slow frontier-model development when necessary.
One line worth unpacking: "A good AI future should not be one where a few institutions control most of the capability and the benefits."
This reflects a pre-IPO posture calibration — in a climate of intense regulatory and public scrutiny over AI safety, the founders are front-loading the "we worry too" language before hitting the road to pitch growth.
Where does OpenAI stand against Anthropic?
Anthropic filed its own confidential IPO application a week earlier, at a valuation of roughly $965 billion — above OpenAI's $852 billion.
The revenue gap is starker: Anthropic's annualized revenue is $47 billion; OpenAI's February figure was $25 billion. Anthropic has built a clear lead in enterprise clients.
This means → The investment-bank logic is blunt: the first company to list will define how investors price the entire AI sector and will absorb capital from funds seeking AI exposure. OpenAI is now the chaser, not the front-runner.
What risks must investors digest?
Burn rate: OpenAI plans massive spending on semiconductors and data centers through 2030. Investor projections call its cash burn unprecedented among public companies, with no positive free cash flow expected before 2030.
Revenue-target shortfall: The Wall Street Journal reported that OpenAI recently missed some internal revenue targets.
Affiliated-company governance risk: On the same day, Altman's co-founded Tools for Humanity — known for the iris-scanning World project (formerly Worldcoin) — reportedly began layoffs. The company raised at a $2.5 billion valuation but now faces revenue struggles, a ban in Kenya, and a $830,000 fine in South Korea.
What is the core suspense in this IPO race?
In plain terms = Altman must answer two questions on the roadshow simultaneously: when can OpenAI's high-burn model sustain itself, and will controversies at his affiliated companies become a governance liability.
ChatGPT has over 900 million weekly active users; a $122 billion funding round closed in March — strong cards, but not the answer.
This reflects the fundamental tension of AI-sector IPOs: valuations rest on a "change the world" narrative, but public-market investors ultimately demand cash flow. Who lists first and sets the pricing benchmark may matter more than whose technology is stronger.
Content is for reference only, not financial advice.