Sequencing an initial public offering ahead of a rival is a genuine strategic move, and the people who advise on these things will tell you exactly why. If you believe investors will prefer your competitor, you file first and capture the available capital before they can. The logic is sound. It is also almost entirely irrelevant to the question enterprise technology leaders need to answer before committing a multi-year platform contract to either Anthropic or OpenAI.

Anthropic filed its confidential S-1 with the U.S. Securities and Exchange Commission on June 1, 2026, roughly one week after closing a $65 billion Series H round at a $965 billion post-money valuation. Its annualized revenue run rate crossed $47 billion in May, up from $9 billion at the end of 2025. The filing puts it marginally ahead of OpenAI, which has been working with bankers on its own offering but had not submitted as of this writing. Both are targeting a public debut as early as this fall, with combined capital raises potentially reaching $100 billion.

The prevailing read in financial coverage is that this is a race with a winner: Anthropic filed first, Anthropic wins the narrative. The Lyft-Uber comparison gets made repeatedly, suggesting the latecomer absorbed more capital and suffered less post-IPO pain. What that framing misses is that Lyft and Uber were competing for rideshare revenue, not for the right to be the compute infrastructure your organization runs on for the next decade.

The revenue trajectory is real. So is the burn.

Anthropic's revenue growth from $87 million in annualized run rate in January 2024 to $47 billion in May 2026 is one of the most compressed revenue ramps in enterprise software history. Salesforce took roughly two decades to reach $30 billion in annual revenue. Anthropic crossed that mark in April.

Claude Code is the primary driver. The agentic coding assistant launched publicly in mid-2025 and crossed $1 billion in annualized revenue within six months. By February 2026, it was generating over $2.5 billion in run-rate revenue. Its adoption across enterprise development teams accelerated a valuation reversal: Anthropic at $965 billion post-money now tops OpenAI's $852 billion mark from late March.

But disclosed alongside that revenue figure was a compute cost that reframes the entire picture. Anthropic's deal with xAI, disclosed in SpaceX's own S-1 filing, costs Anthropic $1.25 billion per month. Annualized, that is $15 billion in compute spend from a single infrastructure partner. As Anthropic co-founder Daniela Amodei told Bloomberg: training frontier models and serving inference at scale requires capital that private markets can only provide so much of. Public markets are the logical next source.

That is what the IPO is. Not a liquidity event. A capital facility.

"If they get out first, they are going to take a lot of the available IPO capital with them, and that is a strategic move, because dollars matter when it comes to these two players." Jeff Bernstein, capital markets adviser, Riveron

What the filing sequence tells enterprise buyers

The assumption in most financial coverage is that the audience watching this race is primarily investors. For enterprise technology leaders, the IPO sequencing is a secondary signal. The primary signal is what public market accountability does to the product roadmap and the pricing model of the platform you are building on.

Both Anthropic and OpenAI are spending capital at a scale that has no precedent in enterprise software. OpenAI's annualized revenue has crossed $30 billion. Both companies are unprofitable. Both are competing for the same pool of compute capacity. A confidential S-1 filing opens a regulatory review window with the SEC. It does not guarantee a listing date, a pricing, or post-listing operational stability.

The historical comparisons in rival IPO coverage favor the company with the stronger underlying business. When Datadog went public after Dynatrace in 2019, it tripled in value at the one-year mark against Dynatrace's double, not because it filed second, but because its product was winning the market. When JD.com went public four months before Alibaba in 2014 and outperformed in the first year, it was despite being far smaller, not because it was first.

In the current AI race, Anthropic has the valuation lead, the revenue lead, and the enterprise focus that public market investors have explicitly said they prefer. OpenAI has the consumer reach with ChatGPT and a rapidly accelerating coding agent, Codex, now past five million weekly active users. Both are credible platforms. Neither is finished being built.

SpaceX changes the capital math for everyone

The context the filing-order narrative largely ignores is SpaceX. Its S-1 is already submitted and its roadshow was underway as Anthropic filed. If Goldman Sachs projections circulated to prospective investors prove directionally accurate, SpaceX will burn roughly $120 billion in 2026 alone. It is the largest IPO in history. The institutional capital it will absorb from the market in the weeks before the Anthropic and OpenAI offerings is not a minor variable.

Enterprise procurement decisions do not wait for public offering calendars. But the stability of the platforms those decisions lock organizations into does depend, partially, on whether the underlying companies can continue funding the compute capacity that makes the models run.

Whether Anthropic or OpenAI lists first is a question for bankers. Whether either company's business model can sustain the capital intensity required to remain a frontier model provider through a public market cycle, that is the question a CIO or CTO should be sitting with right now.