The AI race is moving into budgets, and Anthropic has the narrow lead
Canonicalized generated TECH&SPACE image asset📷 AI-generated image / TECH&SPACE
- ★Anthropic has overtaken OpenAI in Ramp’s U.S. B2B spending index.
- ★The index measures spending, not real usage or long-term customer loyalty.
- ★OpenAI still has room to reset the ratio through pricing, products, and distribution.
According to the source material, for the first time, Anthropic has surpassed OpenAI in B2B adoption, according to Ramp’s spending data—a metric that tracks enterprise AI expenditures rather than raw user counts. The numbers are tight: 34.4% of US companies on the Ramp AI Index now favor Anthropic, edging out OpenAI’s 32.3%.
The shift is striking given Anthropic’s trajectory: its B2B reach quadrupled in a year, while OpenAI’s growth flatlined at 0.3%. Yet the data’s limitations are glaring. Ramp’s index measures spending, not actual usage, meaning the gap could reflect pricing strategies or bundled deals rather than true preference.
The Decoder’s report hints at three factors that could erode Anthropic’s lead, though it stops short of naming them. Speculation centers on cost pressures, model performance, and OpenAI’s potential counter-moves—like pricing adjustments or feature rollouts. What’s clear is that the AI market is far from settled. As one industry observer put it, "We’ve never seen a software industry this dynamic, where newcomers can disrupt leaders in months."
Ramp’s index shows a spending flip, not proof of durable dominance
Canonicalized generated TECH&SPACE image asset📷 AI-generated image / TECH&SPACE
The source material also shows that the real question isn’t who’s ahead today, but whether Anthropic can hold the lead. OpenAI’s stagnant growth suggests internal challenges—perhaps related to model fatigue or enterprise skepticism—but the company’s resources and ecosystem remain formidable. Meanwhile, Anthropic’s rapid ascent could be a flash in the pan if its models fail to deliver consistent value or if costs spiral.
Ramp’s data, while useful, offers a narrow lens. It excludes non-US markets and ignores open-source alternatives gaining traction in Europe and Asia. The index also doesn’t account for multi-model adoption, where companies might use both Anthropic and OpenAI for different use cases. This blind spot matters: if enterprises are hedging their bets, the spending gap could collapse as quickly as it formed.
For now, the numbers are a snapshot of a market in flux. The bigger story is the erosion of vendor stickiness—a trend that could upend the AI landscape before the year is out. The only certainty? Next quarter’s Ramp report will tell a different tale.
