The Data That Changed the AI Business Landscape
New figures from the fintech firm Ramp have reshaped what we know about enterprise AI adoption. For the first time, Anthropic claims more paying business customers than OpenAI. This reversal signals a meaningful shift in how organizations choose their AI providers. It is not a narrow lead, but it is the first time Anthropic has held the top position in this specific metric.
The data comes from Ramp’s AI Index, which tracks expense data across more than 50,000 companies that use the platform. That sample size makes the findings noteworthy even if the methodology has limits. The index captures real spending, not just trial usage or free tier activity. When a business pays for Anthropic or OpenAI services, Ramp sees it. That distinction matters because talk about AI adoption often outpaces actual paid commitment.
For anyone involved in technology procurement, this shift raises important questions. Should your company follow the trend? What drove this change in the first place? And how durable is Anthropic’s new position? The answers require a closer look at the numbers and the strategy behind them.
Fact 1: Anthropic Holds the Top Spot for the First Time
According to Ramp’s survey data, 34.4% of participating businesses now pay for Anthropic services. That figure edges past OpenAI’s 32.3% share. While the gap is modest, the direction of travel matters more than the margin. Anthropic has been gaining ground among high-adoption verticals like finance, technology, and professional services for some time. Ramp economist Ara Kharazian confirmed that Anthropic already led within those groups. The new development is that OpenAI’s remaining lead across other firm types has been shrinking steadily over recent months.
Consider what this means for a procurement manager evaluating AI tools. If your company operates in tech or finance, the trend is especially relevant. Those sectors adopted Anthropic early and continue to increase their spending. But the data suggests the pattern is spreading to other industries as well. The lead is no longer confined to early adopter verticals.
For a startup founder deciding between building on Anthropic or OpenAI’s API, this information carries weight. Platform choice affects development timelines, model behavior, and long-term costs. Knowing which provider has stronger business momentum can inform that decision, though it should not be the only factor.
Fact 2: The Twelve-Month Growth Gap Is Striking
The past year has been transformative for Anthropic’s business segment. In May 2025, only 9% of businesses in Ramp’s sample paid for Anthropic products. That figure climbed by 26 percentage points over the following twelve months. During that same period, OpenAI’s share declined by 1%. The overall share of businesses using some kind of AI product increased by 9%, so the market grew overall. Anthropic captured most of that new spending while also taking share from its rival.
This growth trajectory tells a compelling story. A 26% gain in paid business customers within a single year is unusual in enterprise software. Most B2B AI tools see gradual adoption curves shaped by budget cycles and compliance reviews. Anthropic’s acceleration suggests something different is happening. Either the product itself created strong pull, or the company’s go-to-market strategy unlocked demand that competitors missed.
The answer likely involves both factors. Anthropic’s models earned a reputation for reliability and safety, which appeals to businesses that need predictable outputs. At the same time, the company made deliberate choices about which customers to pursue first and how to expand from there.
What Drove the Surge in Anthropic Business Customers
Understanding why anthropic business customers grew so quickly requires looking beyond the raw numbers. The strategy behind the growth reveals lessons for any company watching the AI market. It also helps explain whether the shift is likely to continue or reverse.
Fact 3: Starting Technical, Then Broadening Out
Kharazian described Anthropic’s approach in clear terms. The company started with a very technical customer base. It focused on their specific needs, succeeded in execution, and then began broadening out through tools like Cowork. That sequence sounds simple, but it runs counter to how many AI companies operate. The temptation is to chase the largest possible market from day one. Anthropic chose a narrower path.
Technical users are demanding customers. They notice inconsistencies in model behavior. They push for lower latency and better reasoning. They are quick to abandon tools that do not meet their standards. By winning this group first, Anthropic built a product that could satisfy high standards before attempting broader adoption. The technical customer base also generated detailed feedback that shaped product improvements.
Once the product was solid, Anthropic could expand to less technical segments. Tools like Cowork lowered the barrier for teams that did not want to write prompts or manage API calls directly. This pattern of diffusion from niche to mainstream is well documented in technology adoption. It worked for cloud infrastructure, for developer tools, and now it appears to be working for AI services.
For a CTO at a mid-size firm, this strategy has practical implications. A provider that earned its growth through technical excellence may offer more reliable long-term support than one that grew through marketing alone. The product was tested by demanding users before it reached your team.
Fact 4: The Data Comes from a Credible but Imperfect Source
Ramp’s AI Index draws from expense data across more than 50,000 companies. That sample includes businesses of various sizes and industries, making it broad enough to carry weight. However, the index only represents companies that use Ramp for expense management. That introduces selection bias. Companies on Ramp tend to be more financially organized and may skew toward certain sectors.
Despite this limitation, the general trend appears across other data sources. On OpenRouter’s leaderboard, which samples a different user population, OpenAI last ranked above Anthropic in December 2025. Since then, Anthropic has held the top position. Two independent data sets telling the same story strengthens the case that a real shift is underway.
For a venture capitalist tracking AI market dynamics, this convergence of signals is significant. When multiple data sources point in the same direction, the trend is more likely to be genuine rather than a statistical artifact. The pattern suggests that Anthropic’s gains reflect real changes in customer preference, not just quirks in Ramp’s sample.
The reliability question also matters for product managers integrating AI features. If you are choosing a vendor based on market momentum, you want confidence that the momentum is real. Cross-referencing Ramp’s data with OpenRouter’s leaderboard and other industry reports provides that confidence.
What Comes Next for Anthropic Business Customers and the Market
The current numbers favor Anthropic, but markets shift quickly. Anyone making decisions based on these trends should consider what comes next. The sustainability of Anthropic’s lead depends on factors that are still unfolding.
Fact 5: Skepticism About Whether the Lead Will Last
Kharazian himself expressed doubt about whether Anthropic’s advantage will persist. He noted that the past year proved the company chose a good strategy and executed it well. But he stopped short of predicting continued dominance. His caution reflects the reality of the AI industry. Competitive advantages can erode quickly when new models launch or pricing changes.
Several factors could reverse the current trend. OpenAI could release a model that significantly outperforms Anthropic’s offerings on key benchmarks. Pricing changes from either company could shift buying behavior. Enterprise customers might consolidate around a single provider for simplicity, and that provider might not be Anthropic. The overall market is still young, and customer loyalty is not yet deeply entrenched.
You may also enjoy reading: Unitree Debuts $650K China Robot Juggernaut Gundam.
For a procurement manager evaluating AI tools, this uncertainty suggests a diversified approach. Committing entirely to one provider carries risk. Building workflows that can switch between models or use multiple providers offers flexibility. The data shows Anthropic is strong today, but the market could look different in another twelve months.
For a startup founder, the lesson is about timing and strategy. Anthropic’s success came from a clear sequence: serve technical users first, execute well, then expand. That blueprint can apply beyond AI. Any company building for business customers can learn from the focus and patience that defined Anthropic’s approach.
Questions to Ask Before Making a Decision Based on Market Share
Market share data is useful, but it should not drive every decision. Before switching providers or committing to a platform based on these trends, consider a few practical questions.
First, does the provider meet your specific technical requirements? A general lead in business customers does not guarantee that a model performs well for your particular use case. Test the actual outputs against your data and workflows.
Second, what is the total cost of ownership? Anthropic’s pricing structure may differ from OpenAI’s. Factor in API costs, integration expenses, and any training your team needs. A market leader might not be the most cost-effective choice for your situation.
Third, how portable are your integrations? If you build deeply into one provider’s ecosystem, switching later becomes expensive. Design your architecture to allow model swapping. This protects you if the market shifts again.
Fourth, what is the provider’s track record on reliability and uptime? Business customers care about consistency. Check service level agreements and look for reports of outages or performance degradation. A growing customer base can strain infrastructure, so monitor how well the provider scales.
Finally, consider your team’s expertise. If your developers are already familiar with one platform’s API, the switching cost includes retraining. Balance the potential benefits of following the market trend against the real cost of change.
The Broader Implications for Enterprise AI Adoption
The shift in anthropic business customers reflects a larger pattern in enterprise AI adoption. Companies are moving from experimentation to committed spending. The 9% overall increase in businesses using AI products confirms that the market is growing, not just reshuffling.
This maturation brings new challenges. Procurement teams need evaluation frameworks that go beyond hype. IT departments need governance policies for AI usage. Finance teams need to track AI spending as a distinct category. The Ramp data shows that this tracking is already happening, and the numbers reveal real preferences.
The fragmentation of the market also suggests that no single AI provider will dominate enterprise spending the way some predicted. Businesses are choosing based on specific needs rather than brand recognition. That trend benefits providers with strong technical foundations, like Anthropic, but it also keeps the door open for new entrants.
For anyone watching this space, the next twelve months will be telling. Can Anthropic maintain its growth rate as it expands to less technical customer segments? Will OpenAI respond with competitive pricing or a breakthrough model? How will other players like Google and Meta position themselves in the enterprise market? The answers will shape the AI landscape for years to come.
What remains clear is that the era of automatic OpenAI dominance in business is over. Anthropic proved that a focused strategy, technical excellence, and patient execution can win paying customers. Whether that lead endures or fades, the lesson for the industry is already written.
