Resources:
Join the discussion on LinkedIn: Got something to say? Let us know on LinkedIn and network with other AI leaders
Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup
Connect with Jordan Wilson: LinkedIn Profile
Start Here Series in our Inner Circle Community: Join for free access
Enterprise AI Adoption: Why Business Leaders Should Question Headlines About Anthropic and OpenAI
A flurry of headlines recently claimed Anthropic has surpassed OpenAI in enterprise AI adoption, with stories proliferating across major outlets such as Business Insider, TechCrunch, and The Wall Street Journal. However, a closer examination of the underlying study at the heart of these claims reveals critical flaws in methodology and interpretation. Business owners and decision-makers evaluating AI providers should look far beyond the surface of viral reports.
AI Study Methodology: Ramp's Metrics Do Not Equal True Enterprise Adoption
Analysts and journalists have widely referenced a study published by Ramp, a spend management fintech platform, which claimed Anthropic had overtaken OpenAI in business AI adoption among its customers. The Ramp AI Index reported Anthropic with 34.4% “adoption” versus OpenAI at 32%, based solely on payment transactions observed among its client base of approximately 50,000 US businesses.
However, Ramp’s dataset doesn’t account for the number of seats, true usage volume, or employee engagement on these platforms. The metric counted positive transaction amounts for an AI product within a single month as de facto "enterprise adoption"—meaning a single purchase by any employee, regardless of company size, counted as enterprise-wide adoption. This binary, yes-or-no framework misses the true depth and breadth of AI adoption in medium and large businesses.
Comparing Anthropic and OpenAI: The Real Numbers Tell a Different Story
Independent of the Ramp study, the best available metrics give a vastly different perspective on relative market positions:
OpenAI publicly reports 9 million paying business users
Anthropic last confirmed 300,000 business customers
While nuances like API spend are not captured in these figures, the differential—roughly 30 times more for OpenAI—is impossible to ignore. Attempts to equate or invert the stature of either provider based on a narrow payments dataset mislead enterprise buyers.
Relative growth rates in the Ramp study further underscore misunderstanding of data basics. Where Ramp claimed a 3.8% increase, simple math on their numbers actually shows 12.5% relative growth—an error stemming from conflating percentage points with percent growth. Decision-makers relying on these metrics should proceed only with a full understanding of these discrepancies.
Microsoft AI and Google AI: Missing Giants in the Data
Ramp’s dataset omits enterprise-scale leaders like Microsoft—an outsized player in AI, with 450 million paid Microsoft 365 seats and reported $37 billion in AI annual recurring revenue. The lack of Microsoft’s presence in the chart alone signals the study’s results are not a proxy for the broader enterprise market. Google, too, appears only marginally in the data, calling into question the representativeness of Ramp’s clientele in reflecting true enterprise AI usage across the Fortune 500 or comparable high-scale organizations.
Incentives and Conflicts: Ramp's Commercial Stakes in AI Spend Narratives
Ramp promotes and markets premium AI spend intelligence tools as part of its business offering. The publication of studies that stoke drama or hype around provider leadership in AI spend can act as de facto sales collateral and PR fodder. Notably, Ramp is publicly referenced by Anthropic as a key customer, has communication ties with Anthropic engineers, and shares a significant overlap in investors with Anthropic’s latest funding rounds—roughly a third compared to less than 15% with OpenAI. This alignment provides financial incentive for headlines equating to enterprise adoption chaos or disruption, which must be considered when reviewing their conclusions.
Baseline: What Reliable Enterprise AI Metrics Look Like
For pragmatic business evaluation, public numbers and search demand provide more meaningful indicators:
OpenAI: 9 million business users, 900 million weekly active users
Anthropic: 300,000 confirmed business users
Monthly search demand: “Claude for business” tallies 720 searches; “ChatGPT for business” exceeds 27,000 searches—a 40x demand difference
Enterprise buyers executing major AI purchasing decisions—commitments sometimes in the millions—should anchor vendor evaluations in these publicly supported facts, not in studies where a single line on a corporate card can be misconstrued as market dominance.
Best Practices in AI Provider Selection
Data quality and methodology scrutiny are non-negotiable. When new studies claim to reveal dramatic market shifts, critical examination of sample size, underlying metrics, and potential funding biases is imperative. Every business needs to look beyond catchy headlines to safeguard both financial outcomes and strategy integrity.
While Anthropic has delivered impressive AI products and OpenAI continues to lead by most quantifiable metrics, business decisions must rely on verifiable, apples-to-apples comparisons. In this environment—marked by pre-IPO hype, rapid product development, and aggressive narrative shaping—only rigorous analysis ensures sound enterprise AI provider choice.
Conclusion:
Enterprise AI adoption is too central to delegate to viral interpretations. Business leaders should demand transparency, interrogate data sources, and ground their investment decisions in facts that reflect actual business use—not just what appears on the top of a payments spreadsheet.
Topics Covered in This Episode:
- Anthropic vs OpenAI Enterprise AI Adoption
- Ramp AI Index Methodology Critique
- Viral Headlines and Media Misrepresentation
- OpenAI vs Anthropic Business User Numbers
- Ramp Study’s Misleading Adoption Metrics
- Missing Microsoft in Enterprise AI Rankings
- Ramp’s Investor and Competitive Overlap
- Category Contradictions in Ramp’s Data
- Real Enterprise AI Adoption Statistics
- Impact of Bad AI Data on Decision-Making
Episode Transcript
Jordan Wilson [00:00:19]:
Meet Firefly AI assistant, now live in Adobe Firefly, the all in one creative AI studio. Just describe what you wanna create, and the assistant handles the rest, orchestrating multi step workflows across Photoshop, Premiere, Express, and more in one conversational interface. You direct the outcome, the assistant accelerates execution. You're gonna be reading a lot of headlines today and probably later this week about how Anthropic is now leading the AI enterprise battle over open AI. And they're all gonna sound super definitive. And matter of fact, so it's gotta be legit, right? Because big name publications ran with this story like business insider and tech crunch, Axios, and even the wall street journal. But should you believe all these headlines, especially when they were all based on the same single study that is misleading at best and failed at middle school math despite positioning itself as a reputable data source? So I don't think the most important topic here is if Anthropic or OpenAI is winning the enterprise. I mean, we'll find that out soon enough as both companies prepare to go public later this year and have to release a lot of their financials.
Jordan Wilson [00:01:43]:
What's important to dissect today is the decisions that your company has to make when it comes to enterprise AI products to use and the data that you use to make those decisions based upon. Because as we'll talk about today, some of the biggest and stickiest data points that companies are making these decisions on are pretty bad and they may not all be true. And let me just get this out of the way. I'm a former journalist and I'm a fan of facts and data. And I'm approaching this look at the ramp study as understanding how the combination of pre IPO marketing and also lazy journalism can lead your company astray if you're not careful. And I'm not saying that anthropics Claude is bad or worse than OpenAI. I'm focusing on facts, misconceptions, and truth that you need to make the right decision for your company. Ready to get into it? Welcome to everyday AI.
Jordan Wilson [00:02:41]:
Before we do double down, let's zoom out and go over the big picture here. A viral chart, I think, is completely misleading enterprise buyers, and the media completely ate it up. Social media completely ate it up. So if you didn't see it, we're gonna go over in more depth today the Ramp AI index. So Ramp is a spend management platform, and they're claiming that Anthropic beat OpenAI in business AI adoption. But just based on their very limited and extremely skewed dataset, that's not even very sound. So the ramp AI index reports went super viral online today, and normally when they release it once a month, it does kinda go viral. But the reason why it went mega viral today, and I think it's gonna continue to grab headlines, you know, throughout the rest of the week is because at least according to their data, it's the first time that Anthropic overtook OpenAI, in the enterprise.
Jordan Wilson [00:03:38]:
And we'll get into their methodology, which I said is bad, here in a couple of minutes. But here's the real facts. Right? All we can do is base this on facts. We've seen all these numbers about anthropics reported revenue, their run rates, open AI is right. It seems like they're kind of neck and neck now. The only thing we actually know until both of these companies go, public and file all their pre IPO paperwork, the only thing we know is OpenAI publicly reports 9,000,000 paying business users. Well, Anthropic last confirmed 300,000 business customers. Not great at math, but that's roughly about 30 x more for OpenAI.
Jordan Wilson [00:04:17]:
I know, obviously, not taking into account API spend and all these other things, but that's the gist. So on today's show, here's what we're gonna uncover. You're gonna learn why ramps and thrapics beat OpenAI headline doesn't match what ramp actually measured and how it actually contradicts some of their own findings. I know confusing. You're gonna see how Microsoft and Google break the entire enterprise AI adoption story that Ramp is trying to sell you and why Ramp has a direct commercial incentive to hype all of this AI spend chaos in provider drama. Alright. Let's get into it. Welcome to Everyday AI.
Jordan Wilson [00:04:55]:
If you're new here, my name is Jordan Wilson, and I do this every day. It's an unedited, unscripted, livestream podcast, and free daily newsletter, helping everyday business leaders like me and you keep up with all of these AI advancements. All of these studies, I tell you, here's the facts. Here's what's matter. Here's what matters, and you take that information to grow your company and your career. It starts here, but make sure you go to youreverydayai.com. That is your cheat code. It is the Gen AI University.
Jordan Wilson [00:05:22]:
You can go now, listen, watch, read about 750 plus, episodes in our catalog, all free, and make sure you go sign up for the free daily newsletter. We're gonna be recapping, today's podcast and a whole lot more. So let me just say this one bluntly. You can probably tell by the title of this episode. No. Anthropic isn't leading in enterprise AI adoption. They're leading in one very bad, piece of data that was put out there, and I will break it down meticulously as I generally do. And, yes, I I hate doing these things over and over.
Jordan Wilson [00:05:55]:
You know, actually broke down a ton of bad studies last year in 2025 that stole the kind of narrative between the MIT study, the MENLO Ventures. So, you know, we have an early candidate for, you know, some of our bad studies for 2026. Alright. But let's look at it. So Ramp, if you don't know, they are a corporate card and spend management fintech platform serving, they say, 50,000 plus businesses. So they published a report, yesterday, that they put out, I think, just about every month titled, Anthropic beats OpenAI on Business Adoption. So the report claims that Anthropic hit about 34.4%, adoption amongst their customers, while OpenAI hit 32%. So, I'll show you the chart here in a second.
Jordan Wilson [00:06:46]:
But here's the thing, and we're gonna be, yeah, going deep onto this one. Their their their metrics and methodology make absolutely no sense. Alright. Let me say this. I'm not an analyst by trade. Alright? But I consult the biggest, company that does analysis. I I do have a slight background in m and e in my, you know, nonprofit days. Just did a lot of m and e work, measurement, evaluate, measurement and evaluation work for Nike and Jordan brand.
Jordan Wilson [00:07:19]:
You know, that's who our nonprofit part, non prop partner, nonprofit partnered with for ten years. Yeah. The thing this thing is really unscripted. Sometimes I say the wrong words. Right. So and I'm I'm a former journalist. Right? I did that earlier in my career. One ACP story of the year.
Jordan Wilson [00:07:36]:
I was a Pulitzer fellow. So, you know, I'm I'm at least, I think, somewhat qualified to look at studies like this from a variety of different perspectives. Well, number one, is this telling the AI story? Yes or no? Number two, is this sound, in terms of its methodology, how the study was conducted? And last but not least, telling you, well, here's what's actually happening in the enterprise. So let's look at the Ramp AI index. So this is, from their, kind of description. I'm not gonna read the whole thing. I have a couple passages highlighted here on my screen. Nothing super visual, but if you ever do wanna see the video version of this, you can go to our website at youreverydayai.com.
Jordan Wilson [00:08:20]:
So here's, some of the snippets describing, their study and their methodology. So they say, our sample includes more than 50,000 American businesses and billions of dollars in corporate spends. Firms are considered to have adopted AI tools if they have a positive transaction amount for an AI product or service in a given month. It's likely our results underestimate actual adoption rates. Alright. There's some things, you know, insert some audible, ellipses between those. Right? So there's, you know, just grabbing some different highlights here. Next, they say, well, ramp customers don't skew toward tech startups.
Jordan Wilson [00:09:01]:
They skew high growth and tech forward. We've observed they are growing faster than the average US business, and they are, after all, adopting a tech forward financial operations platform. Then they also say that I found this one funny. We make no misrepresentations, and we don't make claims if we're not confident about the work that got us there. Okay. Alright. Let's look at this one. We make no misrepresentations.
Jordan Wilson [00:09:30]:
This goes back to my days as a journalist. You know, there's a a a saying, if it bleeds, it leads. Okay. So let's talk here, Ramp, about not making misrepresentations. Your heading here in no shortage of words, says anthropic beats open AI on business adoption. That is the definition of misrepresentation. This there there there's no asterisk on that headline. That headline is well, it's misleading.
Jordan Wilson [00:10:01]:
It's not entirely accurate. It's an opinion. It's based on a very skewed and extremely miniature, dataset. Alright. And then, yeah, we're gonna get into this. This is bad. This is the middle school math, y'all. We're gonna get to this here in a second.
Jordan Wilson [00:10:16]:
Alright. And then in the, the subheading here, it says, today, for the first time, Anthropic passes OpenAI in business adoption according to our latest release of Ramp AI index. Yeah. Put that in the headline, not in the subhead. Right? No one reads this this small print. It says adoption of Anthropic rose 3.8% in April to 34.4% of businesses. OpenAI adoption fell 2.9% to 32.3%. Overall, AI adoption rose 0.2 percentage points to 50.6.
Jordan Wilson [00:10:49]:
Alright. And then here's the chart, right, the chart that you're gonna see a lot. Gosh. Especially if you're on Twitter. FYI, go look at all the people who are sharing this, and you're gonna find out a lot of these people, they are anthropic ambassadors, just FYI. Beside the point, they can share this if they want. Right? That's their own prerogative, but you always need to be, like, look into who's sharing this and look into, well, what their motive may be. Alright.
Jordan Wilson [00:11:13]:
Anyways, here is the, the ramp, AI index here that shows slowly over time. Right? So if you look back, you know, to January 2025 as an example, OpenAI, according to their data, had an overwhelming lead. And now for the first time in April 2026, their data shows in tropic overtaking OpenAI for the lead. Alright. And then obviously, once this came out, it went viral. And I think it's gonna continue to go viral, especially as OpenAI and Anthropic are now, you know, no longer quarters away. They are months away, from filing their pre IPO paperwork. So again, from being in journalism and understanding public relations and spin communication, there's gonna be a lot of stories like this.
Jordan Wilson [00:12:05]:
And, you know, kind of this also, I think, goes to the, you know, the mythos is, you know, the anthropic Claude mythos is too powerful to release, narrative, you know, yet, you know, on benchmarks, their unreleased model is now getting passed by, you know, GPT five five Pro. Microsoft came out in, Beats Anthropics, Mythos in Cyber Gym, even though Mythos is not available. And they said, you know, cyber is its main capabilities. And I'm sure that, you know, Google is probably gonna be announcing a new model at IO next week. That's probably gonna overtake mythos on some of these benchmarks. Anyways, so much of what you have to what you're seeing now coming out of INTROPIC, way more than OpenAI is is usually just spin communication. And it's no surprise to me that as soon as this ramp study hits, all of the brass at anthropic were tweeting about this and saying, oh, we can't keep up with enterprise demand. Right? But I I I don't know.
Jordan Wilson [00:13:06]:
It doesn't look to me like ramp companies are true enterprise companies. I'm trying to see what percentage of, you you know, ramp customers are Fortune 500 as an example. Right? OpenAI says that 95% of the Fortune 500 uses OpenAI's products. So I was trying to see, like, oh, what percentage? Is it 90 per you know, you don't see any of that. Right? It seems like they have a large customer base, but generally just kind of these tech Ford, quote, unquote tech Ford companies. But, yeah, this thing is going viral, online. Everyone's talking about it, the news, and you're gonna continue to hear about it. So let's go over, nine reasons why.
Jordan Wilson [00:13:41]:
I cut two out. Let's go over nine reasons why you probably should ignore this, quote, unquote, study. But first, quick word from our partners. Adobe just introduced an entirely new way to create, bringing the power and precision of its creative suite into one conversational experience. Meet Firefly AI assistant, now live in the Adobe Firefly app, the all in one creative AI studio. Powered by Adobe's creative agent, Firefly AI Assistant lets you start with your vision, just describe what you want, and shape the outcome as it takes form with the assistant. The assistant orchestrates multi step workflows, drawing on 60 plus pro grade tools across Adobe Creative Cloud apps, including Photoshop, Illustrator Premiere, Lightroom Express, and more to help bring your ideas to life. You can also get started with creative skills, a growing library of pre built workflows for common creative tasks like batch editing photos, creating mood boards, portrait retouching, and creating social variations.
Jordan Wilson [00:14:43]:
Every step the assistant takes is visible, so you can refine, redirect, or take over at any time. You stay in the driver's seat as the creative director. Adobe Firefly AI assistant now in public beta. See it today at firefly.adobe.com. Alright. Ready. Reason number one, Ramp is measuring payments, not adoption. This is huge.
Jordan Wilson [00:15:09]:
All they're doing is they're saying, okay. This showed up in our system. So they're not measuring the number of employees using these. They're not measuring seats. They're not you measuring usage. They're not measuring deployment. They're not technically measuring adoption. They're measuring something on a balance sheet.
Jordan Wilson [00:15:28]:
If a company pops up yes or no, that's what they're measuring. You know, they're measuring whether a business had a paid transaction that ramp could see. So calling that business adoption, well, that's just where the, quote unquote study starts to really distort. Alright. Number two. My gosh. All right. And I'm not trying to be mean here again.
Jordan Wilson [00:15:54]:
I'm a big fan of facts. All right. I'm not, I'm not the one putting out a study, Right? Obviously on the podcast, this is live unedited to do. Do I say things that could maybe be disproven down the line? Absolutely. If I'm putting out a quote unquote reputable study that I know each time I put it out, it's going to get picked up by the media. I am going to emphasize truthfulness, making sure my numbers check out this thing. Never should have seen the print. Right? Let's just talk about some middle school math here.
Jordan Wilson [00:16:31]:
Alright. Thanks, missus Janice. I learned this in seventh grade. Alright. The difference between percentage and percentage points. How can anyone take this seriously when ramp itself seemingly doesn't know the difference between percentage and percentage points. Alright. I'm not literally not making this up yet.
Jordan Wilson [00:16:54]:
So they said, alright. Let me let me bring this up here, and and and get my my my notes here as well. So they said, Anthropic past OpenAI in business adoption according to our latest release of the Ramp AI index. Adoption of Anthropic rose 3.8. No, it didn't. It rose according to their data, 12.5%. So ramp putting out this study and journalists wrote about it. Did they not read it? It took me four seconds and that's not an exaggeration.
Jordan Wilson [00:17:34]:
It took me four seconds to instantly know that the numbers here literally don't add up because I read that and I'm like, okay. I I'm pretty okay at percentages in my head, and, you you know, I'm I'm doing the math, and I'm like, no. That's, like, 12%. Right? It was 12.5%. Same same thing with, you you know, OpenAI. If if you don't know the difference between, percent and percentage points, Probably shouldn't be putting out something that you're marketing as something reputable. Just saying. Alright.
Jordan Wilson [00:18:10]:
Sorry. I mean, sorry, not sorry. I think it's important. Right? So, yeah, that's 12.5% relative growth, not up, you know, 3.8% or what whatever they said. Absolutely terrible. If you don't know the difference between percentage points and percent, don't put out a study. Yeah. And I did download the data.
Jordan Wilson [00:18:36]:
There's no data here. Right? I mean, I mean, there's a spreadsheet. I've never seen anything that gets quoted so much in the media. Right? Even going back to the terrible, you know, MIT 95% of, you know, AI pilots fail, at least they had 40 pages of bad information supporting it. This ramp I I and I'm not exaggerating. Their methodology, we had more content every single day in our newsletter. Their methodology is a a couple paragraphs. Right? Their their data points, you can go download it.
Jordan Wilson [00:19:12]:
It's a couple lines in a spreadsheet. There's, like, no real information. You can't actually choose look into the methodology anywhere. It's anyways, I I did confirm that. Yeah. It's it wasn't 3.8% increase. Right? It's it was, 12.5% relative growth. Facts are important.
Jordan Wilson [00:19:33]:
Alright. Next. And this is bigger than all of them, I think. Number three is the doozy. Alright. One employee in theory, and this is just technically an assumption, but, Ramb, if I'm wrong here, call me out. But it doesn't look like I am. Essentially, this is a yes or no.
Jordan Wilson [00:19:54]:
It's binary. It's zero or one. Right? So one employee buying Claude can would just count, right, if it shows up on their ramp card. Right? Because, of course, you know, PWC that has a 100,000, you you know, Chad GPT enterprise seats is is definitely putting that on their ramp card. Right? Yeah. No. No. They're not.
Jordan Wilson [00:20:17]:
No one's putting these things on their card, FYI. You you know, serious serious companies. I've trained companies that have tens of thousands of seats both in Claude, and in, Chat GPT. No one's putting these things in their ramp card. First of all, clients I'm talking to, real enterprise clients, they've never heard of ramp, number one. Number two, yeah, people aren't using their credit card for this. Anyways, according to my interpretation of the data, and I could be wrong here, but I literally read the entire methodology. And like I said, there's a couple paragraphs, and it doesn't explain it.
Jordan Wilson [00:20:51]:
But the way that I read it and interpret it, is that one employee buying Claude can count as a whole company. So it's the same thing. Whether me, everyday AI, right, I am checked. Right? If I use Ramp and I for my, you know, my Claude team account or whatever, you know, it's gonna count the exact same as someone, company with a 100,000. Right? PWC as, you know, 100,000, seats if they use their ramp car. Right? That's all it is. It's it's their customers, and it's a percentage of those customers, which makes absolutely no sense. Like, literally, that if if this is true, that example then probably plays out many times.
Jordan Wilson [00:21:37]:
Right? And my assumption and, again, going back to the 30 x, the last reported number of business customers we got from Anthropic was 300,000, and the last we got from OpenAI was 9,000,000. Right? The assumption has always been that anthropic initially has been developer friendly. So smaller shops, tech forward shops. Right? So it makes sense that, you know, that is ramps crowd, so to speak. But this this is useless. This if if if this is true, which I think it is, the entire basis, let alone that they can't get their math right and don't know the difference between percentage and percentage points. But the the the how you measured it, the the the methodology is is bonkers. And the fact that journalists wrote about this is maybe even more bonkers.
Jordan Wilson [00:22:26]:
I don't know if I'm mad at ramp or if I'm just more disappointed in the journalist that probably didn't read this or think about it think about it creatively. Right? One thing I always love to do is open up these things in Claude. You you know, I did this. Open it up in Claude 47. Open it up in GPT five five pro. Open it up in, you know, Gemini Deepthink. Give no context. Make sure you do a temporary chat and just say, hey.
Jordan Wilson [00:22:49]:
Is this study valid? Yes or no. Right? They all said no, be because it doesn't make any sense. You know, Ramp isn't measuring adoption. They're measuring yes or no to their current clientele that's already skewed to fit the anthropic, right, tech forward, that's your, for the most part, anthropic because it's people who are a little more dev heavy in 2025. Alright. Here's the next one that makes absolutely no sense. Let's go back real quick. Let's go back to this chart here.
Jordan Wilson [00:23:18]:
Alright. Is there someone you don't notice on this chart? Right. We have anthropic in the lead, then OpenAI, then Google, then x AI, then DeepSeek. Is there a name you don't see on there? The original AI enterprise leader. Microsoft's name is not on there. Microsoft has $450,000,000 paid Microsoft three sixty five seats and a reported $37,000,000,000 plus AI annual recurring revenue. Weird. Weird.
Jordan Wilson [00:23:55]:
How does Microsoft not make the list here? Right? If that doesn't tell you that the audience is just well, anthrop right? This quote unquote study should have just, or could for all intents and purposes have been just conducted outside of anthropics dev day last week. Right? If Microsoft doesn't make the list for enterprise AI adoption, I mean, come on. I can't be the only one, like, chuckling at that. Like, it makes no sense. Right? Absolutely absent. Right. And Google barely makes the chart. So, yeah, any enterprise AI chart talking about enterprise AI adoption, missing Microsoft, you just made it a laughingstock.
Jordan Wilson [00:24:40]:
Number five, ramps sample is not the market. I already alluded to this. Right? So they said they use 50,000 plus customers. That's that's a sliver. Right? To call that, like, boldly to say this is enterprise adoption, we're not talking about 50,000,000 businesses. We're not talking about 5,000,000. We're not talking about 500,000. Talking about 50,000.
Jordan Wilson [00:25:04]:
Right? It's a very small sliver. It's not a small sliver to put out a, this is ramp enterprise AI adoption. That's fine. I'm good with that. To try to paint this as a picture of actual enterprise adoption, not just your sliver, your corner of the internet. It's it's asinine to me. Right? Also, there's there's other data that contradicts theirs. You know, there's some federal census data in terms of looking at overall, you know, AI utilization.
Jordan Wilson [00:25:40]:
I'm not gonna get too much into that because it's already, you know, silly enough. So I actually thought of an analogy for this one. Right? So anthropics or sorry, ramps sampling here. Right. And trying to spin it the way that they're spinning it. This would be like me going to, I don't know, a neighborhood. I'm from Chicago. So let's say I go to the West Loop.
Jordan Wilson [00:26:04]:
Right? And I do a study and I say, alright. And I talk to every single person in the West Loop. Right? There's probably 50,000 people. And and I say, who's the best NFL team? Right. Overwhelmingly, probably gonna get the bears as the 98%. Right. And then I put that out, but in the headline, I don't say anything about West Loop residents, Chicago residents. Again, I used to write, help write headlines for a re write for for a living.
Jordan Wilson [00:26:35]:
And if I just spin this as well, the bears are the best team in the NFL. And then all of a sudden some blogs pick it up and, oh my gosh, the bears are the most popular team. Right? And then sports radio, and then it's on ESPN all of a sudden. Right? I'd be laughed at. So why are actual enterprise buyers why is the media why are people looking at this ramp? I think it's human laziness. I don't think humans they see a study. They see a headlines that, you know, they see someone, you know, random influencer who, you know, is tweeting it out, and they're like, oh, it must be legit. Do do people not read things anymore? Am I the like, sometimes I feel can someone email me? Can someone message me on LinkedIn and and tell me, like, did you read this study or do you just read the headlines? If you just read the headlines, like, that's fine.
Jordan Wilson [00:27:25]:
But sometimes I feel like I'm absolutely going crazy when, you you know, I see all these things take off and I say, okay. Like, let me look at it. And, you you know, it's obviously flimsy. Flimsy at best. Yeah. Anyways, there's my Chicago Bears store. Would would you see that headline and just turn? And be like, Interesting. The bears are the best by far.
Jordan Wilson [00:27:47]:
No. They're terrible. I love the bears, but they're terrible. Right? Same thing with this site. Alright. Number six, Ramp has a business reason to hype this. So Ramp sells AI spend intelligence, right, as part of their premium offering that they charge per seat. So when you can, you know, get a splashy AI spend story out there, you know, that obviously makes Ramp in their spend intelligence tool look like, oh, well, we should probably be buying this.
Jordan Wilson [00:28:16]:
So this study for them not only doubles as market research as great PR, but it's also sales collateral. So I always say, who funded the study, and what do they have to gain for, from it monetarily? Seven, ramps, anthropic optics are kinda bad here. Alright. So, anthropic publicly features Ramp as a Claude code customer and reference account. Anthropic says that Ramp had direct communication channels with anthropic engineers. Ramp's main competitor, Brex, has a very public partnership with OpenAI. And here's the one, and I'm not gonna dive too deep into this because I don't wanna stir the pot too much, but I did do the math here. Nine of 30 named ramp investors overlap with Infropix latest series g.
Jordan Wilson [00:29:07]:
Interesting. Right? And and and granted. Right? There's some overlap with OpenAI investors as well, but I was seeing it was more like, 12%, not a third. So roughly a third of ramp investors and Anthropic's latest, investors overlap. Right? So I don't know. I'm putting myself in the shoes of one of those, anthropic investors who's also a ramp investor. And I could see how a study like this would, well, it would make your valuation go kaboomy. Right? Cause if you look, if you look, so just let's, let's look at a timeline, right? Real quick.
Jordan Wilson [00:29:51]:
So late twenty twenty five. Right? So right before this ramp index started, which I believe it started in January 2026. So let's look at late twenty twenty five. Anthropic's valuation was, like, a $180,000,000,000. Right? Then the ramp index, AI index came out. It showed, oh, anthropic growing each month. It's growing. It's catching open AI.
Jordan Wilson [00:30:14]:
Right? And that started to get picked up heavily on social media. It got a lot of play in mainstream media even as the ramp AI index showed that anthropic was gaining. It got talked about a ton in the the the venture capital private equity circles, and now their anthropic's latest valuation is more than $900,000,000,000. And, am I saying that they're, you know, that the this evaluation, skyrocketing was due to Ramp's AI index? Absolutely not. Right? Anthropic has released some world changing products that are absolutely great. Did the Ramp AI index influence the valuation every step of the way? I'd say absolutely. Right? I've been following it. I've heard people talk about this in casual conversation.
Jordan Wilson [00:31:01]:
VC piece, VC and PE folks talk about this as if it's absolute truth, not knowing you know, understanding the methodology. Alright. Number eight. Their own data picks different winners. Yes. Alright. So ramps AI index that just came out says anthropic beats OpenAI, alright, in their monthly metric. But they have something else called the ramp rate, and it still ranks OpenAI first in generative AI category adoption.
Jordan Wilson [00:31:30]:
So which is it? Our AMP's AI index says anthropic wins in an option, but ramp rate says OpenAI is first in category adoption. Yes. Really. It's here. So it says top ranking generative AI software in May 2026. It's OpenAI. OpenAI was the most used vendor in the generative AI space as of May 2026, making it the top choice. Right? So pretty pretty confusing.
Jordan Wilson [00:31:57]:
Right? I I I guess maybe it's in how they name it. I have no clue, but it seems like ramps even own data is contradicting itself. Alright? And then last but not last but not least, number nine. There's no audit, no error bars, no margins, no broad counts, nothing. Right? Again, I'm not kidding. I read their methodology unless there's another one out there. It's literally shorter than our daily newsletter. I wanna see who these 50,000 businesses are.
Jordan Wilson [00:32:28]:
I wanna see the actual, false positives. Right? What do they count as a transaction? What do they not? Why don't they count seats? Why don't they count revenue? Right. I know that, you you know, certain things in their dataset is, anonymized, but why is there no seat waiting, spend waiting, employee waiting, stable panel audit? Right? All of those things that you would look for in a any standard study that doesn't hold potentially this much weight. Right? The these types of studies move markets, and they they are going to greatly impact the public perception of these companies that are both gonna be valued in Anthropic and OpenAI. By the time they go public, they're both gonna be valued at over $1,000,000,000,000 on the market cap side. Right? These types of studies are making probably, I would say, at least tens of billions of dollars in swing. Right. But you don't even get any methodology yet.
Jordan Wilson [00:33:32]:
People are treating this as truth. All right. So let's end with some facts and stats. Here's what you have. 300,000 business customers for Anthropic, 9,000,000 business users for OpenAI. Those are self reported by the companies. Revenue, who knows what it is. Right? Some of the latest reports now say Anthropic is ahead, but those are, for the most part, anonymous sources.
Jordan Wilson [00:33:57]:
Right? Or, you know, the the the company's hinting, oh, yeah. Now we're at $2,000,000,000. Right? But even with if you look at revenue, Anthropic reports it differently than OpenAI. They're not, you know, essentially, reporting their, you know, gross, you you know, they're reporting it differently. So even the revenue numbers, a lot of it's, you know, anonymous sources or, you know, comparing apples versus, you you know, banana pudding. So there it's not good comparisons. The best comparisons we have right now until those numbers get filed. 300,000, business customers for Anthropic, 9,000,000 business users for OpenAI.
Jordan Wilson [00:34:35]:
And then you can look at other metrics. Right? I'm like, okay. What else is there, like, an actual fact metric on? You can look at things like search volume. Right? Or what are people actually looking for? What are people searching for out there? Right? So I did a poll on that using some Google Trends, Keywords Everywhere, data. So Claude for business, there's about 720 searches for that a month. Chad GBT for business, about 27,000. Alright. So, again, not great at math, but that is, let me do the math on there.
Jordan Wilson [00:35:07]:
So that's, 10 for, like, 40 x. Right? Just about 40 x more, people, demand, people searching for chat GBT for business versus quad for business. And that kinda checks because, well, at least right now, there's 30 x, more paying customers. So that kinda checks. So for me, right, if I'm looking at the facts and the stats, let alone OpenAI has 900,000,000 weekly active users, and we don't know how many Anthropic has. I don't know. I would say OpenAI is probably still winning the enterprise. I could be wrong, but that doesn't matter.
Jordan Wilson [00:35:47]:
Like I said, this is not ultimately about who's winning the enterprise and who's not. It's about you. Cause I feel your pain. I feel the anxiety. I feel the, the pressure. Right? I know who's listening to our show. You are making decisions like this. Are we going to have that? You know, are we going to commit to that $1,000,000 spend with anthropic to get, you know, to roll this out across our enterprise? Right? Should we, you know, upgrade from, you know, 10,000 to 20,000 chat GBT enterprise seats? These decisions are not easy because there's a lot of misinformation, disinformation, marketing spin, pre IPO narrative control out there.
Jordan Wilson [00:36:33]:
All I'm trying to do is give you the facts and stats as I see them. And that's what I'm gonna continue to do. So I hope this show was helpful, but it's maybe I'm just getting tired as the former journalist in me. I see these things take off, and then I read these studies and I say, I am just so disappointed. Right. And I don't want you to be confused or have bad information out there. If nothing else. Now I think I hope instead of just reading this headline, you'll actually go read the ramp study.
Jordan Wilson [00:37:03]:
Maybe you'll read it and be like, oh, this is great. This checks out. Cool. I just want you to be aware. Alright? So I hope this was helpful. If if it is, please take a minute to subscribe on the podcast. I'd appreciate that. If you are listening on Apple Podcasts.
Jordan Wilson [00:37:18]:
If you're listening on Spotify, take a minute to follow the show. Leave us a rating there. That would really help us help more business users like you who just want the no BS straightforward approach. So you can get more of that if that's up your alley on our website at youreverydayai.com. So thank you for tuning in. Hope to see you back tomorrow and everyday for more everyday AI. Thanks y'all. Meet Firefly AI assistant now live in Adobe Firefly, the all in one creative AI studio.
Jordan Wilson [00:37:50]:
Just describe what you want to create in your own words, and the assistant handles the rest, orchestrating multistep workflows across Adobe Creative Cloud apps, including Photoshop, Premiere, Express, and more in one conversational interface. You direct the outcome while the assistant accelerates execution. Stay in control with the ability to step in and refine at any time. See it today at firefly.adobe.com.