For the first time, more businesses are spending on Claude than on ChatGPT. The headline is dramatic. The real story, and what you should do about it, is more useful.
TL;DR
In April 2026, Anthropic’s Claude overtook OpenAI in business AI adoption for the first time, reaching 34.4% of business spend versus OpenAI’s 32.3%, according to the Ramp AI Index. The shift is driven mostly by coding and developer workflows, not chat. For most non-technical teams, the practical takeaway is not to switch tools in a panic, but to standardize on one assistant, learn it well, and re-check that choice every quarter.
Here is a sentence almost nobody expected to write a year ago: more businesses are now spending money on Anthropic’s Claude than on OpenAI’s ChatGPT. [1]
In its May 2026 release, the Ramp AI Index showed Anthropic’s share of business AI spend climbing to 34.4% in April, while OpenAI’s slipped to 32.3%. [1] It is the first time Anthropic has been in front. [2] The gap is small. The direction is not. Anthropic has roughly quadrupled its business adoption over the past year, while OpenAI’s barely moved. [2]
If you run a team and you have been quietly wondering whether you backed the right horse, this is the moment that question stops being abstract. So let’s be clear about what actually happened, what it doesn’t mean, and what you should do with the information.
First, a reality check, because the headline is doing a lot of work. The Ramp AI Index tracks card and bill-pay spending across Ramp’s business customers. [1] That is a real, large, current signal. It is not the whole market.
Ramp’s customer base skews toward US-based mid-market and growth-stage companies. [1] So this is “share of business spend among a particular slice of American companies,” not “share of all AI usage on earth.” OpenAI still has far more total users when you count the hundreds of millions of people using free ChatGPT. [5]
The single biggest driver is boring and specific: code. Anthropic’s Claude Code has become the fastest-growing product in the company’s history, and Claude now commands somewhere between 42% and 54% of enterprise spending on AI coding tools, versus around 21% for OpenAI. [3]
That sounds like a developer story that has nothing to do with you if you work in HR or finance. It matters more than it looks. When engineering teams standardize on Claude, the rest of the company tends to follow, because IT has already approved it, security has already reviewed it, and there’s already a subscription to extend. Tools spread sideways through an org once one department commits.
The second driver is positioning. Anthropic has spent two years selling itself to cautious buyers: enterprises in finance, law, healthcare, and consulting that care about safety, predictability, and not embarrassing themselves. PwC, for example, is rolling Claude out across its workforce. [4] When a firm that size makes that call, it becomes a reference every other risk-averse buyer points to.
Almost nothing about the underlying experience changes for you this week. Claude and ChatGPT are still both excellent at the things most professionals actually do: drafting, summarizing, rewriting, analyzing a messy document, turning rough notes into something presentable.
What changes is the strategic backdrop. If your company is choosing one assistant to put in front of everyone, the momentum, the security reviews, and the “everyone else is doing it” comfort have shifted toward Claude. That’s useful to know when you’re in the meeting where this gets decided.
If you want a plain-English refresher on what Claude’s latest model is good and bad at before that meeting, our breakdown of Claude Opus 4.6 for business professionals covers it without the jargon.
Probably not in a hurry. Here’s the part most hot takes skip: the gap between the top models is now small enough that your skill with a tool matters more than the tool itself. A person who knows how to brief Claude well will out-produce someone fumbling with whichever model topped this month’s chart.
Use this quick test before you change anything:
This same logic applies to the wider “AI is disappointing us” debate. We dug into why some companies get results and others don’t in what the 52% are doing differently, and the pattern is consistent: outcomes track how teams use the tool, not which logo is on it.
Forget the leaderboard for a second. Here is the move that pays off regardless of who’s ahead in June: pick one assistant and get genuinely good at it.
To make a fast, defensible choice, run the same real task through both Claude and ChatGPT this week and compare. Use a prompt like this:
Copy this into both tools
“You are helping me with [my actual job]. Here is a real task I do often: [paste a recent email, report request, or document you had to produce]. Do it now, then tell me what extra context would have made your answer better.”
Whichever one’s output you’d actually send with the least editing is your tool. Commit to it for a quarter. Build a small library of prompts that work. If you want a head start on that, our guide to how AI agents are reshaping the workforce shows where this is all heading and which skills hold their value.
A few things keep this from being a clean “Anthropic wins” story. The lead is narrow and built on one data set. Coding spend is volatile, and OpenAI has the consumer reach, the brand recognition, and far more capital to fight back. [3] Analysts covering the shift are explicit that a few product moves could erase the gap quickly. [3]
So treat this as a snapshot, not a verdict. The smart posture for a non-technical leader isn’t loyalty to a vendor. It’s a quarterly habit: re-check which tool serves your team best, keep your prompts portable, and never let a logo do your thinking for you.
Did Anthropic really overtake OpenAI?
In business adoption, yes, according to the May 2026 Ramp AI Index, which showed Anthropic at 34.4% of business AI spend versus OpenAI at 32.3%. This measures spending among Ramp’s business customers, not total global users, where OpenAI is still larger.
Why are companies choosing Claude over ChatGPT?
The biggest driver is coding and developer workflows, where Claude holds an estimated 42% to 54% of enterprise spend. Anthropic has also positioned itself heavily toward cautious enterprise buyers in finance, law, and consulting who prioritize safety and predictability.
Should I switch from ChatGPT to Claude?
Usually not in a rush. The leading models are now close in quality for everyday office work, so your skill with a tool matters more than the brand. Switch only if your company standardizes on one or if a specific feature clearly fits your job better.
Is Claude better than ChatGPT for non-technical work?
For heavy writing, summarizing, and long-document analysis, many people prefer Claude. For image generation, voice, and the widest range of integrations, ChatGPT often wins. For most drafting and analysis tasks, the difference is small.
Will this ranking change again?
Very likely. The lead is narrow and based on one data set that skews toward US mid-market companies. Analysts note that a few product launches from either company could flip the order, so it is best treated as a snapshot rather than a final result.
About this guide
This article explains the May 2026 Ramp AI Index finding for non-technical professionals and gives a simple framework for choosing an AI vendor. It is written by Sana Mian, who has trained 2,000+ professionals to use tools like ChatGPT and Claude in real work.
Sources
Weekly tips. Real examples. Practical help for busy professionals.
We care about your data, check out our privacy policy.
