Enterprise spending on generative AI has surged to 37 billion dollars in 2025, making it the fastest-scaling software category in history. The technology now captures 6% of the global software market just three years after ChatGPT’s launch.
Tim Tully and Matt Murphy report for Menlo Ventures in their third annual State of Generative AI in the Enterprise report. The venture capital firm surveyed nearly 500 U.S. enterprise decision-makers and analyzed spending across model APIs, infrastructure, and applications.
The spending represents a 3.2-fold increase from 11.5 billion dollars in 2024. The largest share, 19 billion dollars, went to user-facing applications that leverage underlying AI models. At least 10 products now generate over 1 billion dollars in annual recurring revenue, while 50 products exceed 100 million dollars.
The data challenges recent concerns about an AI bubble. An MIT study claiming that 95% of generative AI initiatives fail had rattled markets during the summer. But Menlo Ventures’ research shows broad adoption and real revenue across industries.
Enterprise behavior reveals strong confidence in AI solutions. Once an organization commits to exploring an AI solution, 47% of deals reach production, compared to just 25% for traditional software. This elevated conversion reflects clear immediate value.
A significant shift has occurred in how enterprises acquire AI technology. In 2024, companies were split between building and buying solutions. Today, 76% of AI use cases are purchased rather than built internally. Ready-made AI solutions reach production more quickly and demonstrate immediate value.
Product-led growth drives adoption at unprecedented rates. Individual users now account for 27% of all AI application spending, nearly four times the rate in traditional software. When accounting for personal credit card purchases like ChatGPT Plus, this figure may approach 40%.
Coding has emerged as AI’s first killer use case, capturing 4 billion dollars or 55% of departmental AI spending. Half of all developers now use AI coding tools daily, with top organizations reporting 65% adoption. Teams report productivity gains of 15% or more across the software development lifecycle.
At the infrastructure layer, the competitive landscape has shifted dramatically. Anthropic now commands an estimated 40% of enterprise large language model spending, up from 24% last year. OpenAI’s share has fallen to 27% from 50% in 2023, while Google has increased to 21% from 7%.
Anthropic’s dominance stems largely from its performance in coding tasks, where it holds 54% market share. The company has maintained leadership on coding benchmarks for 18 months, starting with Claude Sonnet 3.5 in mid-2024.
Healthcare leads vertical AI adoption with 1.5 billion dollars in spending, more than triple the previous year. Administrative burden and staffing shortages have driven health systems to embrace automation. The ambient scribe market alone reached 600 million dollars.
Startup companies have gained significant ground against established players in AI applications. They now capture 63% of the market, up from 36% last year. In product development and engineering, startups hold 71% market share. In sales, they control 78%.
Looking ahead, Menlo Ventures predicts AI will exceed human performance in daily programming tasks in 2026. The firm expects spending to continue rising despite falling inference costs. Models are also predicted to move to mobile devices as manufacturers ship dedicated low-power GPU compute.
