As of July 31, 2025, Anthropic has surpassed OpenAI to become the leading provider of enterprise large language models (LLMs), capturing 32% of the market share compared to OpenAI's 25%.

In the coding application sector, Anthropic leads with 42% of enterprise market share, significantly outpacing OpenAI, which holds 21%.

Enterprise spending on large language models has surged to $8.4 billion in mid-2025, more than doubling from $3.5 billion reported in late 2024, reflecting strong investment in AI technologies.

Closed-source models now dominate enterprise workloads, powering 87% of them, while the adoption of open-source models has declined from 19% to 13% due to widening performance gaps.

OpenAI faces challenges, including talent retention issues and scrutiny over safety practices, which have hindered its advancements in the enterprise sector.

Google has emerged as a significant player in the enterprise AI sector, now holding a 20% market share with its Gemini models, contributing to a more diverse competitive landscape.

Anthropic's rise in market share is attributed to the successful release of its Claude Sonnet 3.5 model in June 2024 and the Claude Sonnet 3.7 model in February 2025.

Vendor switching among LLM providers is rare, with only 11% of companies changing their model providers over the past year, while 66% upgraded to newer models from their existing vendors.

The rise of Anthropic underscores the importance of addressing specific enterprise needs with tailored AI solutions, shifting the focus from mere computational power to real-world applicability and efficiency.

The report indicates a trend where enterprises are prioritizing models that offer ease of integration and compliance, alongside their capabilities.

Looking ahead, Anthropic plans to advance agentic AI systems in 2025, aiming for models that can autonomously perform complex tasks, with predictions of AI reaching 'country-level' genius capabilities by 2026 or 2027.

Despite Anthropic's current lead in the enterprise sector, concerns about safety cultures in the AI industry persist, highlighting the need for a balance between innovation and risk management.