The news: The enterprise AI market is solidifying around a few dominant providers, with OpenAI leading but Anthropic gaining rapidly, per a new a16z survey of 100 CIOs from Forbes’ Global 2000 list.

OpenAI currently captures 56% of enterprise model spending. It remains the default vendor for large deployments, pilots that scaled, and early agent pilots. That lead reflects first-mover advantage, brand trust, and broad API integration across productivity stacks.

However, the ChatGPT maker’s grip is loosening. Anthropic’s enterprise penetration climbed 25% between May and December 2025, reaching 44%. That means nearly half of surveyed enterprises now use Anthropic’s Claude in some capacity—often alongside OpenAI’s models. 

Why it’s worth watching: CIOs project that by 2026, OpenAI will hold 53% market share, while Anthropic and Google are each expected to account for 18%.

Most large companies are running multiple models in parallel. Four out of five (81%) CIOs now use three or more model families in testing or production, up from 68% a year earlier.

• OpenAI is preferred for horizontal use cases. It’s predominantly deployed for chatbots, internal knowledge search, and customer support. ChatGPT’s ecosystem depth and incumbency inside large deployments keep it the default choice when CIOs prioritize scale and stability.
• Anthropic is advancing in coding, reasoning, and data-heavy workloads. Three-quarters of Anthropic customers had Sonnet 4.5 or Opus 4.5 in production, showing a preference for the latest models in complex tasks.
• Google sits in the middle. It has broad reach and secure cloud access but uneven traction in tasks like coding. Gemini’s recent surge and ubiquity in Google Workspace positions it for continued adoption.

Implications for brands: Companies now move across copilots built into productivity suites, standalone agents, and cloud-based tools—often running on different models from OpenAI, Anthropic, or Google.

The challenge is workflow consistency, which can be managed through centralized guardrails that travel across tools. Brands should assess models by strength, not habit. Measure success at the workflow level while limiting provider changes to keep costs low and maximize ROI. 

This content is part of EMARKETER’s subscription Briefings, where we pair daily updates with data and analysis from forecasts and research reports. Our Briefings prepare you to start your day informed, to provide critical insights in an important meeting, and to understand the context of what’s happening in your industry. Non-clients can click here to get a demo of our full platform and coverage.