[News] Anthropic Eyes $2T IPO Valuation, $518B AI Buildout Spans Amazon, Google, Broadcom, and More

Anthropic’s IPO filing sheds new light on the AI developer’s growing reliance on a small group of technology partners. According to Reuters, citing a copy of the confidential filing, Anthropic routed 47% of its sales to customers last year through Amazon and Alphabet’s Google, up from 32% in 2024 and 11% in 2023. The two companies are also major Anthropic investors, critical suppliers of computing power, and direct rivals in AI.

Anthropic views its cloud partnerships as an advantage, saying Amazon, Google, and Microsoft’s platforms allow Claude to tap their vast sales networks and reach existing customers, as noted by Reuters. However, the company also acknowledged that relying on a limited number of partners and suppliers could create conflicts of interest and adversely affect its access to compute.

Anthropic is seeking a valuation of about $2 trillion and plans to spend hundreds of billions of dollars in the coming years to accelerate its growth, Reuters notes. Its reliance on major technology companies extends beyond sales and investment, with the company making substantial long-term commitments to secure computing capacity.

Anthropic is expected to begin marketing its IPO as early as mid-October, with the listing potentially completed just days before the U.S. midterm elections in November, according to Reuters, citing sources.

$518 Billion AI Infrastructure Buildout

According to a separate Reuters report citing the prospectus, Anthropic expects to spend at least $518 billion over a decade building AI infrastructure with six partners, with about 80% of that amount either non-cancelable or requiring payment regardless of usage. Anthropic said the commitments are necessary because future demand for advanced AI systems is likely to exceed available supply, with AI development expected to be “limited principally by the availability of compute.”

Under long-term infrastructure service obligations, Anthropic plans to spend at least $111.1 billion with Google, $110 billion with Amazon, and $31.4 billion with Microsoft over the next seven to 10 years, regardless of usage, as Reuters indicates. Separately, the filing disclosed about $161.2 billion in largely non-cancelable Broadcom-related equipment lease obligations. Anthropic also has agreements with xAI that could result in up to $84.5 billion in spending on NVIDIA-based computing capacity through 2029, though those commitments are largely cancelable with 90 days’ notice. AMD is also expected to supply more than $20 billion in AI computing capacity.

Beyond AI accelerators, Anthropic has named Micron, Samsung, and SK hynix as strategic infrastructure partners, saying their memory technologies will help support the continued expansion of its compute capacity as demand for Claude grows, according to the company’s Series H announcement.

Read more

• [News] Anthropic CEO Calls for Slower AI Progress but Flags China Risk; OpenAI Reportedly Rules Out 2026 IPO
• [News] Anthropic Reportedly Eyes Custom AI Chip; in Talks with Samsung for 2nm Foundry and Advanced Packaging

(Photo credit: Anthropic)