Field note

Anthropic passed OpenAI in business AI adoption: what it means for AEO

Ramp's AI Index shows Anthropic overtaking OpenAI in US business adoption for the first time in April 2026-34.4% versus 32.3% by spending share. Here is what the data actually measures, why Claude Code drove it, and why it raises the stakes for being cited and recommended inside Claude.

Buffy Editorial2026-07-12 · 4 min read

Anthropic passed OpenAI in US business AI adoption for the first time in April 2026, according to Ramp's AI Index. Ramp reported Anthropic at 34.4% of businesses (up 3.8 points month over month) against OpenAI's 32.3% (down 2.9). A lead built on spending data across tens of thousands of US companies, reported on 13 May 2026. For anyone doing answer engine optimisation, the signal is simple: buyer attention is shifting toward Claude, which raises the stakes for being citable inside it.

What this data is, and isn't. The Ramp AI Index measures paid business adoption, inferred from real corporate-card and expense spending on Ramp's platform, not consumer usage, market share, or citation share. It is one company's transaction data, so it is directional and US-business-skewed. It is corroborated by independent coverage (Axios, VentureBeat, The Decoder, WinBuzzer, all mid-May 2026), which is why it is worth reading, but hedge it as a spending signal, dated to April 2026.

What exactly did Ramp report?

The month-over-month shift, as Ramp reported it for April 2026:

Provider Share of businesses (Apr 2026) Month-over-month change
Anthropic 34.4% +3.8 points
OpenAI 32.3% −2.9 points

Source: Ramp AI Index, April 2026 data, reported 13 May 2026. The reversal is stark against a year earlier: per Ramp's underlying data, in April 2025 OpenAI held roughly 32% of business adoption while Anthropic was under 8%. So Anthropic did not inch ahead. It closed a gap of more than 20 points in about a year.

Why did Anthropic pull ahead?

Ramp and the coverage around it attribute the swing largely to one product: Claude Code, Anthropic's agentic coding tool, described as the fastest-growing product in the company's history. A separate analysis cited alongside the index estimated that around 4% of all public GitHub commits were being authored by Claude Code. Roughly double the share of a month earlier (a third-party estimate, so treat it as illustrative, not audited).

The mechanism matters for AEO. Coding tools put a specific assistant. Claude. In front of technical buyers and their teams every day. Daily working use is exactly what builds the habit of asking that same assistant non-coding questions too, including "which tool should we use for X." That is how a developer tool becomes a discovery surface.

Ramp's data measures where businesses spend, not where consumers chat. It says buyer attention at work is moving toward Claude, which is a reason to make sure Claude can find, cite, and recommend you.

What does this mean for being cited in Claude?

It raises the commercial value of Claude visibility. If more of your buyers work inside Claude, then whether Claude can retrieve, cite, and recommend you stops being a niche concern. Two things follow:

Crucially, this does not mean shift effort away from ChatGPT. Adoption share and citation share are different things: OpenAI still holds enormous consumer reach, and the engines cite different sources from different pools (see how many sources each engine cites). The honest read is add Claude to what you measure and optimise for, weighted by how much your own buyers use it, not swap one engine for another.

Will Anthropic's lead hold?

Unclear, and worth hedging. VentureBeat's coverage flagged that several threats could erase the lead, and a single month of spending data is a fragile basis for a trend. Adoption among businesses is volatile, procurement decisions lag, and a strong product release from any provider can move the numbers again.

The durable takeaway is not "Claude won." It is that business AI adoption is now genuinely multi-engine and moving fast, so betting your visibility on one assistant is riskier than it was. That is the same lesson the corpus keeps reaching: measure presence and citations across every engine your buyers use, and track the trend rather than a single headline month. The discipline in how to measure AI visibility.

How should you act on this?

A short, honest response beats a scramble:

  1. Find out which assistants your buyers actually use: don't assume; ask, and check your own AI-referral data.
  2. Confirm Claude can reach you: that you aren't blocking its crawler path at the CDN and that Brave can discover your pages.
  3. Measure your citation and recommendation share in Claude and ChatGPT side by side, weighted to your buyer mix, over time.

That last step is the loop that closes it. Knowing whether Claude, ChatGPT, and the rest actually surface, cite, and recommend your brand. Tracked daily across every engine rather than spot-checked. Is exactly what Buffy Intel is built to measure.

Frequently asked

Did Anthropic really overtake OpenAI in business adoption?

By one measure, yes. Ramp's AI Index. Based on real spending across tens of thousands of US businesses on Ramp's corporate-card and expense platform. Reported that in April 2026 Anthropic reached 34.4% of businesses (up 3.8 points) while OpenAI fell to 32.3% (down 2.9), the first time Anthropic led. This is paid-business-spending data reported 13 May 2026, not consumer usage or citation share, so read it as one directional signal, corroborated by Axios, VentureBeat, and The Decoder.

What drove Anthropic past OpenAI?

Ramp and press coverage point to a single product: Claude Code, Anthropic's agentic coding tool, described as the fastest-growing product in the company's history. A year earlier the picture was reversed. Per Ramp's data, OpenAI held roughly 32% of business adoption in April 2025 while Anthropic was under 8%. The lead is recent and, as VentureBeat noted, not guaranteed to hold.

Why does business AI adoption matter for AI visibility?

Because the assistants your buyers use at work are the ones that will answer their questions and recommend vendors. If more businesses are paying for and working inside Claude, then being cited and recommended in Claude, which draws its web results largely through Brave Search. Becomes commercially more important, not just a nice-to-have. Adoption tells you where buyer attention is going; it does not tell you your citation share, which you still have to measure.