The race for enterprise AI dominance is far from settled. According to TechCrunch AI, new data from corporate expense management platform Ramp shows OpenAI is starting to claw back market share from Anthropic among U.S. businesses. Until both AI giants get close enough to their planned IPOs to release official financials, this kind of proxy data remains the best indicator of actual market traction.
The Shifting Leaderboard
Ramp processes billions in spending for over 70,000 American companies. Because it is a popular corporate card in Silicon Valley, its customer base skews heavily toward the tech industry, making it a reliable bellwether for early enterprise adoption.
Back in May, Anthropic pulled ahead of OpenAI in this segment, capturing 41% of the market share compared to OpenAI’s 39%. By July, Anthropic expanded that lead to nearly 44%.
But the third quarter is telling a different story. Ramp economist Ara Kharazian notes that OpenAI is currently growing faster than Anthropic among these same business users. While there is still time left in the quarter, the momentum is clearly shifting back toward the ChatGPT maker.
The Myth of AI Loyalty
What stands out here is the extreme volatility of enterprise AI budgets. Investors banking on highly reliable recurring revenue might need to adjust their expectations. Businesses are not locking themselves into a single vendor. Instead, they are willing to jump ship the moment a competing lab releases a superior model or adjusts its pricing structure.
The TechCrunch AI report highlights that developer preference and regulatory constraints heavily influence these swings. When a model hits the sweet spot of performance and price, adoption spikes. Conversely, higher-end models face immediate headwinds when they introduce friction.
For example, Anthropic caused concern among privacy-conscious enterprises when it warned users of a higher-end tier that it must retain their data for 30 days. High costs combined with strict data retention requirements can quickly stall adoption, even for highly capable reasoning models.
An Expanding Market
While OpenAI and Anthropic trade blows for the top spot, the overall market is expanding at a healthy clip. Ramp’s data shows that the percentage of companies paying for AI tools topped 50% in March and climbed to nearly 56% by July. The pie is getting bigger. Both companies are likely growing their raw business revenue regardless of who holds the market share crown on any given month.
Strategic Takeaways for Businesses
What should AI practitioners and business leaders take away from these shifting dynamics?
- Build for flexibility: Design your AI architecture to be model-agnostic. If a better or cheaper model drops next month, you need the technical flexibility to route queries to a new provider without rebuilding your entire software stack.
- Audit data policies: Pay close attention to data retention rules. As Anthropic’s 30-day retention requirement showed, privacy constraints can create major roadblocks. Always verify how your corporate data is stored and used by third-party labs.
- Match the model to the mission: General chatbots and targeted, high-end models serve distinct purposes. Avoid paying a premium for complex capabilities if your specific use case only requires basic text generation.
The enterprise AI market moves at a blistering pace. This data proves that corporate loyalty in the AI sector is entirely dependent on who shipped the best model last week. For AI labs, success requires relentless innovation, because enterprise customers are always ready to pivot.