Anthropic is close to overtaking OpenAI on this measure of AI business spending

Anthropic now accounts for 30.6% of Ramp customers paying for AI products, up 6.3 percentage points since March, narrowing the gap with OpenAI at 35.2%. Ramp says Anthropic is on track to surpass OpenAI in business AI spending within two months and already leads in information, finance and insurance, and personal services. The article also notes a reputational boost after Anthropic resisted Pentagon terms, which appears to have helped adoption among corporate users.

Analysis

The more important signal here is not a brand-vs-brand race; it is that enterprise AI spend is becoming a distribution contest, and the current winner is the provider that is easiest to operationalize inside workflow-heavy, regulated, and VC-backed firms. That favors vendors with stronger developer mindshare and lower friction in code-adjacent use cases, which creates a second-order boost to software tooling ecosystems and cloud inference demand even if model differentiation narrows over time.

Ramp’s customer mix implies AI adoption is still concentrated in faster-moving balance-sheet-light businesses, but the spread into finance and insurance matters because those sectors have the highest willingness to pay for compliance-safe automation once procurement standards are set. If Anthropic continues taking share there, the next leg is likely not app-store buzz but embedded usage through enterprise software vendors and internal copilots, which would be stickier and less visible in consumer metrics.

The contrarian read is that reputational controversy may have accelerated adoption by making Anthropic the "independent" choice versus the incumbent default. That effect can fade, but the more durable driver is product-market fit in code and document workflows; if OpenAI closes the gap on those workflows or bundles aggressively through Microsoft, share gains could stall within one to two quarters. The key risk to the bullish read is that this is still a spending-share metric, not net revenue share, so a few large contract changes could reverse the headline trend quickly.

Key Decisions for Investors

• Stay constructive on RAMP for 1-2 quarters as the picks-and-shovels beneficiary of rising enterprise AI spend; use weakness to add, but size modestly because the market may already be discounting AI mix expansion.
• Long MSFT / short a basket of pure-play AI inference beneficiaries for a 3-6 month horizon if you believe OpenAI distribution through Microsoft eventually reasserts share; this is a hedge against Anthropic momentum fading into bundled enterprise channels.
• Add to META on any pullback if management continues highlighting internal AI adoption and productivity gains; the read-through is that broader enterprise AI normalization supports ad-tech automation and better internal tooling, with limited direct competitive risk from model-share shifts.
• Relative value: long AI software enablers with developer exposure, short slower-moving horizontal SaaS names that lack code-centric workflows; expected payoff is 10-20% outperformance over 6-12 months if Anthropic’s developer lead persists.
• Watch for a reversal trigger: if OpenAI or Microsoft wins a major regulated-sector deployment in the next 1-2 quarters, trim Anthropic-adjacent optimism and fade the share-gain story, since enterprise adoption can rotate quickly once procurement standards are standardized.