Skip to main content
Editorial sketch of a glass-walled office building at night with financial charts visible through windows and a city skyline backdrop
Sounding
SHALLOWS

Anthropic surpasses $19 billion ARR as Claude Code fuels record growth

VERIFIEDConfidence: 80%

What Happened

Speaking at the Morgan Stanley TMT conference, Anthropic CEO Dario Amodei disclosed that the company has surpassed $19 billion in annualized run-rate revenue — more than double the $9 billion figure reported at the close of 2025, according to Bloomberg. The pace of acceleration is striking even by AI industry standards: $6 billion in new annualized revenue was added in February 2026 alone.

Claude Code is the primary engine behind that growth. Anthropic's AI coding assistant has hit $2.5 billion in ARR, and business subscriptions to the product quadrupled between January 1 and early March 2026, per The Decoder. Enterprise customers now represent 80% of Anthropic's total revenue and account for more than half of Claude Code's run rate.

The revenue update follows Anthropic's $30 billion Series G in February 2026, which valued the company at $380 billion post-money.

On February 27 — the same day OpenAI finalized a Pentagon AI contract — Defense Secretary Pete Hegseth designated Anthropic a supply-chain risk, barring defense contractors from doing business with the company. Anthropic called the designation "legally untenable" and has pledged a court challenge.


Why It Matters

The revenue figures reframe what looked like a damaging government setback. When Hegseth's designation landed, the open question was whether a federal supply-chain label would ripple into enterprise procurement — the kind of regulatory uncertainty that can quietly erode a B2B pipeline. The data so far says it has not.

Editorial sketch of a developer at a desk with code on dual monitors and a revenue growth chart on a whiteboard

The $6 billion added in a single month is not simply a large number; it represents a growth rate that, if sustained, would double Anthropic's ARR again before mid-year. That trajectory reflects two structural dynamics operating simultaneously. First, enterprise AI spending is graduating from pilot budgets to operating-line commitments. Anthropic's 80% enterprise revenue share indicates that its customers are treating Claude as core infrastructure, not an experimental tool. Second, AI coding assistants have become the fastest-growing category within that enterprise shift. Claude Code's ascent from zero to $2.5 billion ARR in under a year is consistent with broader category data showing developer tooling as the highest-adoption enterprise AI segment.

The Pentagon feud matters, but it is commercially separate from these dynamics. Anthropic's refusal to permit mass surveillance or autonomous weapons use cost it federal contracts; it has not cost it the commercial enterprise market. Former White House advisor Dean Ball described Hegseth's designation as "attempted corporate murder." The revenue data suggests the attempt has not landed.

The wider competitive picture is sharpening. OpenAI finalized a Pentagon deal on February 27 and is now in early-stage talks to deploy AI across NATO unclassified networks — CEO Sam Altman initially described them as "classified networks" at an internal all-hands meeting before a spokesperson corrected the record. The two companies' diverging government postures are forming a clear strategic fork: one optimized for commercial enterprise scale, one pursuing Western defense infrastructure. Enterprise buyers operating in government-adjacent industries will be navigating that distinction for years.

Newsletter

Stay informed. The best AI coverage, delivered weekly.

Related