Anthropic Selects Banks to Lead IPO Process

Anthropic has selected Morgan Stanley and Goldman Sachs to lead work on an initial public offering, and JPMorgan Chase is also on the deal, Bloomberg reported on June 3, 2026, citing people familiar with the matter (Bloomberg; Bloomberg Law). Bloomberg and PYMNTS report that Anthropic filed confidentially with the SEC after submitting a draft registration statement, and that the company is weighing a possible listing as soon as October, though details could change (Bloomberg; PYMNTS). PYMNTS reports that Anthropic recently raised $65 billion in a Series H that valued the company at $965 billion post-money, and includes the company quote, "This gives us the option to go public after the SEC completes its review," attributed to Anthropic's announcement (PYMNTS).

Editorial analysis - technical context: Public reporting highlights a material third-party compute arrangement disclosed elsewhere, not in Anthropic filings. The Next Web cites SpaceX's S-1 filing, which reports that SpaceX provides Anthropic with approximately 325,000 Nvidia chips at a cost of around $1.25 billion per month through May 2029, subject to termination terms in the agreement (The Next Web). If the SpaceX disclosure is accurate, that arrangement represents a substantial recurring vendor cost that any eventual S-1 or public filing would need to disclose as a related-party or material supplier agreement.

The reported bank selection and confidential SEC filing come amid an accelerating race among large AI developers to reach public markets. Bloomberg frames Anthropic's move as part of a broader contest with OpenAI for institutional investor attention and the first-mover valuation benchmark (Bloomberg). PYMNTS documents Anthropic's recent private financing trajectory, noting the company's jump from a $380 billion valuation in February to $965 billion after the Series H (PYMNTS). The Next Web highlights that SpaceX's computing-disclosure and overlap in AI products complicate both competitive framing and disclosure obligations for an S-1 (The Next Web).

For practitioners: Observers will look for the public S-1 for three categories of detail that change how practitioners and investors evaluate AI companies. First, revenue and margin breakdowns for model licensing, cloud services, and bespoke enterprise contracts. Second, vendor and related-party agreements for compute capacity, notably the terms reported in SpaceX's S-1 and their impact on cost structure. Third, governance and public-company controls, including how a public Anthropic would disclose valuation reconciliation from late-stage private rounds to public-market metrics. These are standard disclosure items but are especially salient for large, capital-intensive AI firms.