Enterprise AI market widens

Anthropic is closing the gap with OpenAI in U.S. business AI spending, and the fight is shifting from software features to who can secure enough chips and data-center power. (ramp.com) (anthropic.com) Ramp said on April 11 that 50.4% of businesses on its platform paid for artificial intelligence tools in March, up from 35% a year earlier. In the same month, Anthropic’s adoption rate rose to 30.6% of businesses, while OpenAI stood at 35.2%. (ramp.com) That narrowed the gap to 4.6 percentage points from 11 points in February, according to Ramp’s corporate card and invoiced-payment data. Ramp also said Anthropic already leads in information, finance, and professional services, three sectors where AI use is already high. (ramp.com) For companies buying AI, that means the market is looking less like a one-vendor standard and more like a cloud-style procurement contest. Anthropic said April 6 that Claude is available on Amazon Web Services Bedrock, Google Cloud Vertex AI, and Microsoft Azure Foundry. (anthropic.com) The bottleneck is not only model quality. Training and running large AI systems requires vast clusters of specialized chips, the same way a power-hungry factory needs land, electricity, and grid access before it can add production lines. (aboutamazon.com) (openai.com) Anthropic said it signed a new agreement with Google and Broadcom for multiple gigawatts of next-generation Tensor Processing Unit capacity starting in 2027. The company said most of that new compute will be in the United States and tied it to a November 2025 pledge to invest $50 billion in American computing infrastructure. (anthropic.com) Amazon Web Services has already built out one of those giant clusters for Anthropic. Amazon said Project Rainier came online in October 2025 with nearly half a million Trainium2 chips, and Anthropic was expected to scale to more than 1 million chips by the end of that year. (aboutamazon.com) OpenAI is pursuing the same problem from a different angle. OpenAI and SoftBank said in January 2025 that the Stargate Project planned to invest $500 billion over four years in United States AI infrastructure, with $100 billion slated for immediate deployment. (openai.com) (group.softbank/en/news/press/20250122) Amazon has also deepened its financial tie to Anthropic while keeping the cloud relationship central. Anthropic said in November 2024 that Amazon added a new $4 billion investment, bringing Amazon’s total commitment to $8 billion and making Amazon Web Services its primary cloud and training partner. (anthropic.com) Anthropic’s own customer numbers show why suppliers are racing to lock in capacity. The company said April 6 that its revenue run rate had passed $30 billion and that the number of business customers spending more than $1 million a year had doubled to more than 1,000 in less than two months. (anthropic.com) The result is a market where enterprise buyers can increasingly choose among models, but the vendors still need the same scarce ingredients underneath: chips, electricity, and cloud contracts. Ramp’s March data suggest more companies are now paying for AI; the next question is which providers can keep serving that demand at scale. (ramp.com) (anthropic.com)