The Case for Public Capital

Anthropic has filed confidentially for an IPO following a $65 billion fundraise at a $965 billion valuation. Co-founder Daniela Amodei justifies the move by citing the extreme capital intensity of training and serving frontier AI models. She argues that public markets are the most suitable vehicle for the sustained, large-scale investment required to remain competitive in the AI sector.

Strategic Compute Management

Unlike competitors such as OpenAI and xAI, Anthropic has avoided building its own data centers. Amodei describes this as a deliberate strategy to avoid overextending capital on infrastructure that may not be fully utilized. The company prefers to maintain a slight deficit in compute capacity—ensuring demand outstrips supply—rather than risking the inefficiency of idle hardware. This approach is exemplified by their $1.25 billion-per-month compute partnership with xAI, which allows them to scale capacity without the long-term operational burden of physical data center ownership.

Despite rapid growth—with annualized revenue reaching $47 billion in May 2026, up from $9 billion at the end of 2025—the company faces questions regarding the long-term ROI of AI for enterprise customers. Amodei remains optimistic, framing current corporate hesitation as a natural phase of adoption. She believes that as businesses move beyond initial experimentation into deeper integration within coding, healthcare, and financial services, the realized value of these tools will increase, justifying continued enterprise spending.