OpenAI filed a confidential S-1 registration statement with the U.S. Securities and Exchange Commission on June 8, 2026, kicking off the process for what could be the largest technology IPO in history. The company is targeting a valuation that analysts expect to exceed $1 trillion, up from the $852 billion post-money mark it was last privately valued at. A public listing is planned for as early as September 2026.

Goldman Sachs and Morgan Stanley are managing the offering, with JPMorgan also reportedly involved. OpenAI said it chose to publicly announce the filing because it expected the document to leak.

The news came one week after Anthropic filed its own confidential S-1 at a $965 billion valuation. For the first time, both of the AI industry’s dominant model providers are simultaneously preparing for public markets.

A Trillion-Dollar Company That Loses Money

The filing sets up a fascinating test for public market investors. By almost any measure, OpenAI is one of the most consequential companies in the world right now. ChatGPT reached 1 billion users in 2026. The company has enterprise contracts with major corporations across every industry. Its Codex product is expanding aggressively into the developer tools market.

But the financials tell a complicated story. According to analysis of OpenAI’s recent results, the company loses approximately $1.22 for every dollar it earns. The gap between revenue growth and compute costs has narrowed compared to 2024, but OpenAI has not yet reached profitability, and the path to getting there requires either significant price increases, dramatic cost reductions in inference, or both.

Public markets will demand answers to questions that have been fuzzy for years: how much does it cost to serve ChatGPT at a billion users, what margins does OpenAI generate on enterprise contracts, and when does the underlying model business actually become self-sustaining?

None of that is unknowable — it is just currently unknown. The S-1, whenever it becomes public, will be one of the most closely read corporate documents in recent memory.

The Enterprise Codex Bet

The filing comes as OpenAI is making a deliberate pivot toward its enterprise revenue. Codex, the company’s AI coding assistant, has expanded its enterprise offering significantly in 2026, and it represents a higher-margin business than consumer ChatGPT subscriptions.

The logic is straightforward: enterprise software contracts are stickier, more predictable, and more defensible than consumer subscriptions. If OpenAI can shift a meaningful portion of its revenue mix toward enterprise, the path to profitability becomes cleaner.

This is the same playbook Anthropic has been running with Claude’s enterprise tier, and it is the same market that Microsoft (through Copilot), Google (through Gemini for Workspace), and Amazon (through Bedrock) are all contesting.

The IPO effectively forces OpenAI to commit to this enterprise narrative in writing, with full financial disclosure, for the first time.

Two IPOs in One Week

The coincidence of timing between OpenAI and Anthropic is not entirely a coincidence. Both companies have been signaling for months that a public listing was the likely next step. What changed in June 2026 is that both apparently reached the same conclusion at roughly the same time: the market is ready, the revenue is large enough to support a prospectus, and waiting longer carries more risk than going now.

The competitive pressure between them is real. Sam Altman has previously told investors he intends for OpenAI to go public before Anthropic. Anthropic filed June 1. OpenAI filed June 8. Whether that ordering was deliberate or circumstantial, the market now has two competing AI platform IPOs to evaluate side by side.

That comparison will be instructive. Anthropic generates roughly double OpenAI’s annualised recurring revenue relative to capital raised, and has projected positive cash flow by 2028 compared to OpenAI’s 2030 projection. Investors evaluating both prospectuses will need to decide which financial profile they prefer.

What This Means for Business

Your AI vendor is about to have shareholders. Public companies make decisions differently from private ones. Quarterly earnings pressure, margin requirements, and customer concentration analysis will all influence how OpenAI prices its products and prioritises its roadmap. That is not inherently bad, but it is different. Plan for it.

Pricing stability is not guaranteed. OpenAI loses money per dollar earned right now. A public offering doesn’t change that overnight, and the pressure to reach profitability will eventually find its way into pricing. Businesses that have built workflows assuming current API pricing should be thinking about how exposed they are to rate changes.

The era of opaque AI financials is ending. When both Anthropic and OpenAI have public S-1 filings, we will finally know what enterprise AI actually costs to build at scale. That transparency will benefit businesses making vendor decisions. The noise about which model is “better” will start to be supplemented by real data about which provider is financially durable.

Diversification is good strategy right now. Two of the three dominant AI providers are simultaneously navigating IPO transitions. The third, Google, is a publicly traded company already dealing with its own AI investment scrutiny. For businesses that run critical workflows on any of these platforms, now is a good time to ensure you are not single-vendor dependent.

If you want to understand how to build AI systems that are resilient to vendor dynamics while delivering genuine business value, start with a conversation with our team.