Anthropic Overtakes OpenAI in Business Customer Count as Enterprise AI Adoption Hits 47.6%

For the first time since major enterprise adoption of generative AI began, Anthropic has pulled ahead of OpenAI in verified business customer count, according to new data from the Ramp AI Index. The milestone comes as overall business AI adoption reached 47.6% in February 2026, up from 46.8% in January.

The shift reflects a broader competitive realignment in the enterprise AI market. Anthropic captured 37% of trackable business spending on generative AI software in the first quarter, compared to 33% for OpenAI, marking the first time the Claude maker has exceeded OpenAI in dollar terms. Meanwhile, 24.4% of businesses on Ramp's platform were using Anthropic services in February, while OpenAI experienced a 1.5% monthly decline — the largest single-month drop for any AI model company since Ramp began tracking enterprise adoption.

Market Dynamics Shift Toward Anthropic

The customer acquisition patterns reveal significant momentum behind Anthropic's enterprise strategy. Among businesses purchasing AI services for the first time, Anthropic wins approximately 70% of head-to-head matchups against OpenAI. This success rate suggests that when enterprises evaluate both platforms directly, they increasingly favor Claude over GPT models for business applications.

The competitive landscape shows clear sectoral preferences. Anthropic currently leads OpenAI in three specific industries: information services, finance and insurance, and personal services. These sectors represent some of the most compliance-sensitive and risk-aware segments of the enterprise market, areas where Anthropic's emphasis on AI safety and constitutional AI approaches may resonate with procurement and risk management teams.

Despite Anthropic's gains, the relationship between the two companies in enterprise accounts remains complex. Approximately 79% of Anthropic customers also maintain OpenAI subscriptions, while 16% of all businesses tracked by Ramp pay for both services simultaneously. This dual-sourcing pattern reflects the current state of enterprise AI strategy, where organizations hedge their bets across multiple model providers rather than committing to a single platform.

Enterprise Adoption Patterns and Platform Preferences

The February data shows continued acceleration in overall enterprise AI adoption, with month-over-month growth maintaining pace despite the market's increasing maturity. The 47.6% adoption rate represents nearly half of all businesses tracked by Ramp, a remarkable penetration rate for technology that entered mainstream enterprise consideration less than two years ago.

OpenAI's decline, while notable for its magnitude, does not represent a wholesale enterprise exodus. The company remains the AI model provider used by the most businesses as of March 2026, indicating that while new customer acquisition may be challenging, existing enterprise relationships remain largely intact. The 1.5% monthly decline appears concentrated among newer or smaller accounts rather than representing departures by major enterprise customers.

Looking at spending patterns provides additional context for the competitive dynamics. Anthropic's 37% share of trackable business spending suggests that while OpenAI may retain more total customers, Anthropic is capturing larger contract values or deeper enterprise penetration among its customer base. This spending concentration could indicate either premium pricing for Anthropic's services or more extensive deployment across business units within client organizations.

Technical and Strategic Considerations

The enterprise preference patterns emerging from this data reflect several technical and strategic factors that distinguish the two platforms. Anthropic's constitutional AI approach and emphasis on interpretability align well with enterprise governance requirements, particularly in regulated industries where model behavior must be explainable and auditable.

From a practical deployment perspective, the high percentage of dual-sourcing suggests enterprises are treating generative AI models as interchangeable commodities to some degree, while also recognizing meaningful differences in capability, cost, or compliance features. This behavior mirrors earlier enterprise technology adoption patterns, where organizations maintained multiple vendors for critical infrastructure components to ensure redundancy and leverage.

The industry-specific preferences reveal how different sectors evaluate AI model capabilities. Finance and insurance companies, traditionally conservative in technology adoption, appear to favor Anthropic's approach to model safety and transparency. Information services companies may be responding to Anthropic's focus on factual accuracy and reduced hallucination rates. Personal services companies might value Claude's conversational capabilities and customer interaction features.

We have seen this pattern before, when cloud computing providers competed for enterprise market share in the mid-2010s. Initially, AWS dominated through first-mover advantage and technical sophistication. However, competitors like Microsoft Azure and Google Cloud gained ground by focusing on specific enterprise needs — compliance, integration with existing systems, and industry-specific solutions. The AI model market appears to be following a similar trajectory, with differentiation occurring around enterprise-specific requirements rather than purely technical capabilities.

Implications for Enterprise AI Strategy

The shifting competitive landscape carries immediate implications for enterprise AI procurement and strategy. The data suggests that businesses evaluating AI platforms for the first time should expect Anthropic to present stronger value propositions in head-to-head comparisons, particularly in regulated industries or applications requiring high degrees of interpretability.

For enterprises already deployed on OpenAI infrastructure, the market dynamics indicate they are increasingly in the minority if they have not at least evaluated Anthropic as a secondary or complementary provider. The prevalence of dual-sourcing arrangements suggests that maintaining relationships with multiple AI model providers has become a best practice for risk management and capability optimization.

The spending concentration around Anthropic also indicates that enterprises may be finding more extensive use cases or higher-value applications for Claude than for GPT models. This could reflect either superior performance in specific enterprise workflows or more successful integration with existing business processes.

Looking forward, the competitive dynamics suggest continued fragmentation in the enterprise AI model market, with different providers capturing market share based on specialized capabilities rather than universal dominance. Enterprises should prepare for a multi-vendor AI landscape where platform selection becomes increasingly application-specific rather than organization-wide.