Anthropic selects Morgan Stanley and Goldman Sachs to oversee its IPO.

      TL;DR: Anthropic has chosen Morgan Stanley and Goldman Sachs to lead its IPO, with JPMorgan also participating, aiming for a listing in October. SpaceX's S-1 disclosed that it supplies Anthropic with 325,000 Nvidia chips at a cost of $1.25 billion monthly.

      Anthropic has appointed Morgan Stanley and Goldman Sachs as the lead underwriters for its initial public offering, with JPMorgan Chase also involved, as reported by Bloomberg on Tuesday. The developer of Claude is considering a public listing as early as October, having filed for a confidential listing on Monday. Additional banks may be included in the underwriting team, and offering details could evolve.

      This appointment solidifies the underwriting group for what could be one of the largest technology IPOs ever. Anthropic’s listing is part of a series of significant offerings this fall, with SpaceX planning to go public potentially as early as June 12, valued at $1.8 trillion, and OpenAI also preparing its filing with Goldman Sachs, Morgan Stanley, Citigroup, and JPMorgan.

      The notable detail from SpaceX’s IPO filing disclosed that SpaceX provides Anthropic with AI computing resources, including around 325,000 Nvidia chips at $1.25 billion per month. This agreement is set to continue until May 2029, but either party can terminate it with 90 days' notice after an initial three-month period.

      This contract positions SpaceX as an Anthropic supplier, competitor (via its Grok chatbot), and a potential IPO participant. The monthly cost of $1.25 billion translates to $15 billion annually for Anthropic, highlighting the extensive infrastructure needs and the slim margins that a company expecting $10.9 billion in quarterly revenue must manage. The S-1, when released, will have to disclose this as a significant related-party arrangement due to the competitive overlap.

      Anthropic and OpenAI are vying for investor interest, utilizing the same underwriting banks, and ultimately the same pool of public market capital. Both companies have engaged with Morgan Stanley, Goldman Sachs, and JPMorgan. The first company to go public will set the valuation standard for the AI sector and will gain first access to institutional investments.

      Anthropic has significantly improved its position this year. Its Mythos cybersecurity model disturbed global markets, and its coding agents negatively impacted legal software stocks in February. The company's revenue trajectory, projected to grow from $4 billion annually in July 2025 to a $50 billion run rate by July 2026, has outpaced OpenAI's. With a private valuation of $965 billion, Anthropic has surpassed OpenAI’s, altering a hierarchy that existed since both companies were established.

      The Pentagon's risk designation of Anthropic remains a primary concern for the IPO. The company’s refusal to allow the military unrestricted access to its models has led to a rating typically reserved for foreign adversaries, which Anthropic claims could threaten billions in revenue. Investors in the public market will closely analyze how the S-1 addresses and quantifies this risk.

      Founded in 2021 by former OpenAI employees, including CEO Dario Amodei, Anthropic presents itself as a more responsible AI steward compared to its rivals. Claude and its technology have earned popularity among enterprise customers in finance, healthcare, and software development. The surge in enterprise AI spending has disproportionately benefited Anthropic, which is on track for its first profitable quarter, projecting $559 million in operating profit from $10.9 billion in Q2 revenue.

      The IPO will assess whether a company that has raised $65 billion in private funding, achieved a $965 billion valuation, and incurs $1.25 billion monthly on a single computing contract can maintain public investor enthusiasm. Morgan Stanley and Goldman Sachs believe it can.