Menlo Ventures estimates enterprise AI deployment spending reached $176.8 billion

Menlo Ventures says enterprise AI deployment spending rose from $33 billion in 2025 to $177 billion in 2026, a 5.4-fold increase.[2] The firm reports that spending reached $176.8 billion by September 2026.[2] Its report says 55% of AI spending came from new budgets rather than existing software or labor budgets.[2] The survey covered nearly 400 decision-makers at U.S. enterprises responsible for more than $4 billion in AI spending, and found average spending was 4.7 times the previous year’s level.[3]

Menlo Ventures says frontier labs captured 69% of enterprise deployment dollars, compared with 50% the previous year.[2] It estimates that Anthropic, OpenAI and Google together command $122.7 billion in enterprise spending, or 2.3 times the rest of the market combined.[2] Within enterprise API spending, Anthropic’s share rose from 40% to 55%, and OpenAI’s from 27% to 31%; Google’s fell from 21% to 9%.[2] The report also says open-source models accounted for 19% of AI usage but 4% of spending, a difference between usage share and dollars spent.[2]

The report points to substantial agent activity: Menlo Ventures says agents accounted for 80% of tokens on OpenRouter, about seven times human-driven usage.[2] It says agent traffic on the platform grew 37-fold after surpassing human traffic in February.[2] In its survey, 74% of respondents said they used agents for workflows and tasks, while 32% used multiple agents at once.[3] Menlo Ventures also says each dollar spent on developing models generated $1.70 in revenue.[2] These measures cover distinct things—platform token traffic, reported enterprise use and model-development revenue—so they should not be treated as interchangeable.

Menlo Ventures puts the AI app market at $77.3 billion and says coding and chatbot products together made up 76% of it.[2] It estimates the coding AI market at $34 billion and the chatbot AI market at $25 billion.[2] The firm says coding is 96% AI-native and estimates Claude Code and Codex together account for $24 billion in coding spend.[2] It also says coding agents from AI labs grew from $1 billion to $24 billion, while the rest of the coding market grew 2.9-fold to $10 billion.[2] The report says 22% of enterprise software budgets were redirected toward AI, and 13% of that redirected money still went to traditional software companies for new AI products.[3]