OpenAI submitted a confidential S-1 to the SEC on Monday, marking the first major step toward an OpenAI IPO. The company announced the filing in a blog post, saying it expects the confidential filing to leak.
The move comes just over a week after Anthropic filed to go public, per TechCrunch, and sets up a 2026 IPO season that’s shaping up to include two of the three largest AI labs plus SpaceX at a combined valuation north of $3 trillion.
But the real story isn’t the filing itself. It’s what OpenAI published alongside it.
Let me break down what happened, what it means, and why the AGI manifesto might matter more than the S-1.
The Filing: What OpenAI’s IPO Paperwork Actually Says
OpenAI’s announcement is refreshingly direct. “We recently submitted a confidential S-1. We expect it to leak, so we’re just announcing it.” The company said it hasn’t decided on timing, and that it may stay private longer if it makes sense for what it wants to build.
That last bit is important. OpenAI just published a curated Codex use-case hub and rebuilt ChatGPT’s memory system in the same week they filed. They’re shipping product, building developer relationships, and telling the IPO story all at once.
OpenAI was valued at $852 billion in March, when it raised $3 billion from retail investors. But it also recently missed its own targets for new users and revenue. Going public now is a choice, and the company itself acknowledges the tradeoff: some things are easier as a private company.
The filing is under Rule 135 of the Securities Act, which means it’s a placeholder. No details on share count, price range, or exchange. That all comes later if and when OpenAI decides to actually list.
The Anthropic Context
Anthropic filed its own confidential S-1 around June 1. The same company that won a court injunction against the Trump administration in March is now racing OpenAI to public markets. The two largest AI labs are in a race to IPO, and the timing matters.
Both companies are unprofitable. Both are burning enormous amounts of capital on compute. Both face an S&P 500 that just refused to bend its rules for SpaceX.
The S&P 500 requires 12 months of seasoning, 10% public float, and four consecutive quarters of profitability. OpenAI would net an estimated $8 billion in passive fund buying if it gets in. Anthropic would net $4.6 billion. Neither is guaranteed entry under current rules.
The S&P 500 decision is the subtext of both filings. SpaceX wanted fast-track entry at its $1.75 trillion IPO valuation. The S&P said no. That door is closed for OpenAI and Anthropic too.
The AGI Manifesto
OpenAI published a sweeping AGI mission statement titled “Built to benefit everyone: our plan” on the same day it filed its OpenAI IPO paperwork. It’s a sweeping philosophical statement about AGI, the company’s mission, and its vision for the future. It talks about giving everyone on Earth a personal AGI, building automated AI researchers, and accelerating the economy.
This is the kind of forward-looking communication that companies entering a quiet period have historically been careful to avoid. OpenAI appears comfortable publishing it right alongside an S-1 filing. That says something about the regulatory environment it’s operating in.
The SEC under the current administration has taken a markedly more hands-off posture toward tech and AI companies. OpenAI’s legal team seems to have judged that publishing a grand vision statement about AGI hours after filing to go public won’t draw scrutiny. They’re probably right.
The timing also serves another purpose. The AGI mission statement frames OpenAI as a mission-driven organization, not just another tech company cashing out. It’s a narrative hedge. “We’re not going public because we need to. We’re going public so we can give AGI to everyone.” Whether you buy that framing or not, it’s a smart preemptive move.
What This Means
Three of the most closely watched companies in tech could all go public within months of each other. OpenAI, Anthropic, and SpaceX represent a concentration of high-stakes offerings the markets haven’t seen since the dot-com boom.
For anyone building on top of these platforms, the changes are coming. Public companies report quarterly earnings. They face shareholder pressure. They optimize for different things than private companies do.
I wrote about how Google ate OpenAI’s lunch in 2025 and the same market dynamics that drove that story will only get more intense when both labs are answering to public shareholders. OpenAI’s API pricing, model release cadence, and product strategy will all be judged by public market standards if and when the IPO happens.
The S-1 is the first step in a long process. The actual IPO could be months or even a year away. But the race is officially on. Two labs filed in the same month. The third isn’t far behind.
I’ll be watching the SEC filings and the quiet period protocols. If OpenAI starts talking less and shipping more, you’ll know why.
Disclosure: News analysis based on published reporting.
