Enterprise AI Adoption Flips in Anthropic’s Favor
Recent spending data shows a decisive shift in enterprise AI adoption. Ramp’s expense platform, which tracks actual corporate card transactions across more than 50,000 companies, now records Anthropic capturing 34.4% of discrete AI purchases, edging past OpenAI at 32.3%. This is not survey sentiment but hard payment data for standalone AI subscriptions. Over just one year, Anthropic’s share jumped from roughly 9% to the mid-30s, while OpenAI’s declined, including a sharp 1.5‑point drop in February, the largest single‑month fall for any tracked AI vendor. Because the index excludes most bundled cloud deals, it highlights which tools businesses actively choose when they are not tied to existing platforms. In that context, Claude’s rise over ChatGPT signals that enterprise AI spending is increasingly driven by targeted workflow needs and perceived reliability rather than pure brand familiarity.
Revenue Surge Positions Anthropic for Its First Profitable Quarter
Anthropic is expected to post its first profitable quarter in Q2, with revenue projected to approach USD 10.9 billion (approx. RM50.1 billion). That level would represent roughly 130% sequential growth from an estimated USD 4.7 billion (approx. RM21.6 billion) in Q1 and would exceed the company’s total revenue for all of last year. Investor materials reviewed by the Wall Street Journal point to a potential USD 559 million (approx. RM2.6 billion) operating profit on that Q2 revenue base. The company has already seen a USD 30 billion (approx. RM138 billion) annualized run rate come into view, and its latest funding round reportedly targets a valuation near USD 900 billion (approx. RM4.13 trillion). Yet Anthropic itself has cautioned that profitability may not be steady throughout 2026 as it continues to ramp up infrastructure and compute spending to keep pace with demand for Claude.
Claude vs ChatGPT: Why Workflows Matter More Than Model Hype
The emerging Claude vs ChatGPT battle in corporate environments is being decided less by headline benchmarks and more by day‑to‑day workflow fit. Anthropic has focused on winning over technical teams first—developers, data scientists, and AI‑literate staff—and then expanding usage across the wider organization. Once Claude proves reliable in complex internal workflows, procurement teams increasingly standardize on it for broader AI business spending. This stands in contrast to more generalist deployments of ChatGPT that often start with experimentation and lightweight use cases. For enterprises, the priority is not just model IQ, but predictable capacity, latency, and guardrails that can support high‑stakes workloads. As companies embed AI deeper into processes like document review, knowledge management, and customer operations, these operational assurances are becoming decisive. Claude’s momentum in discrete purchasing data suggests it is winning where usage translates directly into measurable business productivity.
Funding, Compute, and the Race to Sustainable Margins
Anthropic’s path to early profitability is tightly bound to how it balances revenue growth against rising compute costs. Investor materials indicate compute spending is expected to fall from 71 cents to 56 cents for every dollar of revenue, a major efficiency improvement that underpins the projected USD 559 million (approx. RM2.6 billion) operating profit. At the same time, Anthropic has committed to significant infrastructure expansion, including a USD 1.25 billion (approx. RM5.73 billion) monthly deal with SpaceX for computing power at Colossus data centers through May 2029 and an extended 300‑megawatt footprint. CEO Dario Amodei has admitted that revenue growth has been “too hard to handle,” with the company preparing for 10‑fold growth but experiencing 80‑fold annualized growth in Q1 instead. The central question now is whether enterprise AI adoption will continue to outpace these escalating fixed costs over multiple quarters.
What Anthropic’s Lead Means for the Future of Enterprise AI
Anthropic’s projected profitability years ahead of OpenAI—whose internal forecasts reportedly do not envision profits until 2029 or 2030, with a USD 74 billion (approx. RM339.0 billion) loss expected in 2028—signals a strategic divergence in how AI leaders are approaching scale. Anthropic appears to be building around paying enterprise workloads first and then stretching infrastructure to meet that demand, rather than prioritizing maximal model capability at any cost. For CIOs, this makes Anthropic a test case for whether a foundation model provider can pair cutting‑edge performance with sustainable unit economics. If Claude continues to capture a larger share of AI business spending while maintaining, or regaining, profitability amid heavy compute investment, it will set a new benchmark for enterprise AI adoption. Vendors that fail to show a similar alignment between revenue and infrastructure costs may find it harder to win long‑term, workflow‑critical deployments.