Anthropic Closes $30 Billion Funding Round at $900 Billion Valuation, Overtaking OpenAI

At some point in the week ending May 29, 2026, Anthropic quietly completed the largest private AI financing round ever assembled — and, in doing so, became the most valuable private AI company on earth. The $30 billion-plus round closed at a pre-money valuation exceeding $900 billion, pushing Anthropic past rival OpenAI, which was last valued at $852 billion in March. It is the capstone on a revenue trajectory that has been, by any historical measure, extraordinary.

The Capital Stack

Four firms co-led the round with approximately $2 billion each: Sequoia Capital, Dragoneer Investment Group, Altimeter Capital, and Greenoaks Capital Partners. Existing investors Peter Thiel’s Founders Fund and General Catalyst also participated. The co-lead structure — four firms at equal commitment rather than a single lead — reflects both the scale of the round and the competitive dynamics of growth equity right now, where top-tier crossover funds have been scrambling for allocation in frontier AI companies.

The structure also carries a message about Anthropic’s leverage. The company reportedly received significantly more interest than the $30 billion it sought, enabling it to set terms rather than negotiate them. People familiar with the round say Anthropic did not entertain a lead investor in the traditional sense, instead keeping the co-lead format to avoid concentrating any single firm’s governance influence.

Revenue: A Number That Stops Conversations

The financial case for a $900 billion valuation is built on a revenue trajectory that has few precedents in enterprise software, let alone AI. At the end of 2025, Anthropic’s annualized revenue was approximately $9 billion. By the close of Q1 2026 it had surpassed $30 billion annualized. For Q2 2026, Anthropic has told investors it expects to report $10.9 billion in revenue — more than doubling the prior quarter. If that lands, the annualized run rate at the end of Q2 will be roughly $44 billion.

The projection investors are most focused on, however, is further forward. Anthropic has told prospective investors that its annualized revenue run rate will exceed $50 billion before July 2026. That would represent roughly a 450 percent increase in less than 18 months from a business that was sub-$10 billion annualized as recently as late 2025. Anthropic’s growth rate, if sustained even briefly, would rank among the fastest enterprise revenue ramps in the history of technology.

The Business Behind the Numbers

Anthropic’s revenue growth has three reinforcing drivers, and understanding them separately matters for assessing how durable the trajectory is.

The first is enterprise depth. The number of Anthropic customers spending more than $1 million annually on an annualized basis has reportedly exceeded 1,000, with that count doubling in under two months. Enterprise deals with KPMG (276,000 employees across 138 countries), Amazon (Claude as the default model in Bedrock’s enterprise tier), and multiple defense and intelligence agencies have added large, recurring, contracted revenue blocks that are structurally different from the per-query consumption of earlier developer API usage.

The second is the agentic workload shift. Claude’s usage profile has changed dramatically since the Managed Agents API launched in early 2026. Agentic tasks — where Claude autonomously executes multi-step workflows over hours rather than answering a single query — consume dramatically more tokens per session than conversational use. Anthropic’s per-customer token consumption has risen faster than its customer count, meaning revenue per customer is expanding even without price increases.

The third is developer platform lock-in. Anthropic’s expanding ecosystem of tools — Claude Code, the Model Context Protocol (MCP), and pre-built agentic frameworks — creates integration dependencies that make Claude stickier inside development workflows. Enterprises that have built internal agents on MCP have meaningful switching costs, and that cohort is growing.

Why $900 Billion — and Is It Justified?

The headline number will attract skepticism, and it should be interrogated honestly. A $900 billion pre-money valuation on $44 billion of annualized revenue implies a roughly 20x forward revenue multiple, which is elevated even by high-growth SaaS standards. The conventional counterargument is that Anthropic is not a SaaS company in the traditional sense: it is building a general-purpose intelligence layer with the potential to restructure multiple industries simultaneously, and the total addressable market is not a segment of software spend but a meaningful fraction of global economic activity.

The more immediate risk is margin. Anthropic’s compute costs are staggering — running frontier models at the scale of current usage requires sustained infrastructure spend that has not historically correlated linearly with revenue for AI labs in the way it might for a conventional software business. Anthropic has disclosed that it is in early-stage discussions with Microsoft about running Claude inference workloads on Microsoft’s custom Maia 200 AI chips via Azure, which would improve inference cost efficiency if successful. The company also secured a $40 billion data-center deal through SpaceX’s Colossus facility — a relationship that reflects the infrastructure requirements at scale.

The Positioning Game: Anthropic vs. OpenAI

Surpassing OpenAI in valuation is symbolically significant, but the more important contest is market position in the enterprise and agentic segments that will generate the bulk of AI revenue over the next three to five years. OpenAI’s DeployCo subsidiary, its broader ChatGPT Enterprise business, and its deep integration into Microsoft 365 give it formidable distribution advantages that a valuation number does not erase.

What Anthropic has built that OpenAI has not, at least not as clearly, is a perception of trustworthiness with the large enterprises that are most sensitive to AI liability risk. Anthropic’s Constitutional AI approach, its transparency about model limitations, and its willingness to give enterprise customers fine-grained audit and control tools have resonated particularly strongly in regulated industries — financial services, healthcare, law — where OpenAI’s brand has carried more consumer-product associations.

Whether that positioning advantage holds as OpenAI builds out its own enterprise trust and governance features, and as Google’s Gemini Enterprise gains traction, is one of the central questions of the next eighteen months.

What the Money Buys

Anthropic has been clear internally and externally about its capital priorities: training runs for the next generation of frontier models, infrastructure for the agentic workload surge, and the international expansion required to capture enterprise markets outside the United States. The company opened its Milan office earlier this month; additional European offices are expected before year-end.

The round also provides Anthropic with the balance sheet to absorb compute cost volatility in a market where GPU pricing, cloud contracts, and energy costs remain unpredictable. A startup with $30 billion in fresh capital can negotiate infrastructure contracts from a position of strength that a company with one quarter of runway cannot.

At $900 billion, Anthropic is valued at more than most Fortune 50 companies. The question now is whether the revenue trajectory that justified that number continues — or whether, as has happened repeatedly in AI’s short history, the growth curve inflects in ways that the best models couldn’t predict.

For the moment, Anthropic’s investors are betting heavily that this time, the curve does not bend down.