Anthropic is reportedly continuing preparations for an initial public offering, setting up a potential public-market test for an artificial-intelligence company whose chief executive is simultaneously calling for more restraint around the most capable systems. The company has not announced a listing date, exchange, share price or public prospectus.
CNBC reported on Sept. 12 that Anthropic was actively preparing for an expected IPO. Six days later, New York Times reporter Erin Griffith wrote in a public post that the company was still pursuing an offering while expanding its business. The reporting puts a sharper focus on how investors would value growth, spending and governance at a company that has made safety proposals part of its public policy posture.
IPO preparation is not a public offering
The available evidence supports a narrower conclusion than an imminent deal: Anthropic appears to be preparing for an IPO, but the offering itself remains unannounced. CNBC did not report a timetable or terms, and Anthropic has not publicly disclosed either.
Investing.com reported that Anthropic confidentially submitted a draft Form S-1 registration statement to the Securities and Exchange Commission on June 1. Such draft submissions can remain nonpublic while a company and regulators work through comments. Anthropic has not publicly confirmed the reported confidential submission, and no public registration statement has been filed.
The distinction is consequential for investors. A public filing would ordinarily provide the first standardized look at such matters as historical financial statements, losses or profitability, customer concentration, contractual obligations, risk factors, executive compensation and governance arrangements. None of those documents is yet available through a public prospectus, nor is there a confirmed ticker symbol, exchange or price range.
Financing points to scale, not audited earnings
Anthropic has separately disclosed unusually large private-company figures. In its Series H announcement, the company said it raised $65 billion at a $965 billion post-money valuation and that its annualized revenue run rate had crossed $47 billion earlier that month. The company said the funding was expected to support safety and interpretability research, computing expansion, and the scaling of products and partnerships.
The $65 billion financing equals roughly 6.7% of the stated $965 billion post-money valuation, illustrating the amount of fresh capital associated with the round. But the revenue figure requires a separate reading. Run-rate revenue annualizes a recent level of sales; it is not audited revenue for a completed fiscal year, and it does not establish profit, cash flow or the durability of that sales pace.
That qualification also helps reconcile a lower revenue-run-rate figure cited in the Investing.com analysis, which referenced roughly $30 billion in April. The figures may reflect different points in time rather than conflicting measurements. A public filing, if one emerges, would show whether and how the company presents its historical revenue under public-company disclosure rules.
Safety proposals could become an investor governance issue
The IPO reporting arrives as Dario Amodei, Anthropic’s co-founder and chief executive, has argued that developers of advanced AI should move more deliberately. CNBC reported that his proposed approach includes independent evaluators with employee-level access, coordination among leading companies in democratic countries, and international government coordination.
Amodei’s proposal, as described by CNBC, does not call for ending AI training or technical development. Instead, it argues for enough time for alignment work, safeguards and third-party evaluation before the capabilities of frontier systems advance further. The operational effect of such a framework would depend on how a company defines a high-risk capability, selects evaluators and responds to their findings.
For a future shareholder base, those choices could become concrete governance questions rather than abstract policy positions. Independent evaluators with meaningful access could require disclosure controls and limits on sensitive information. A decision to delay a model release for testing could affect product timing, while a decision to proceed could raise questions about the safeguards used. Neither outcome is established by the current reporting, but a prospectus could clarify the company’s formal risk oversight and the authority assigned to its board and management.
What remains to be disclosed
Anthropic’s private financing announcement supplies a valuation and company-stated operating metrics, while the independent reports establish that IPO preparation is under way. They do not establish the economics of a public deal. Investors still lack basic offering details as well as public financial statements that would allow them to assess the relationship between the company’s reported sales momentum, investment needs and safety commitments.
For now, the clearest public record is a company with a stated $965 billion private valuation, a claimed $47 billion revenue run rate and reported IPO preparations—but without a public registration statement to show how those figures and its approach to frontier-AI risk would be presented to public investors.