ChatGPT now reaches more than 800 million people every week, a milestone that OpenAI says is reshaping how businesses adopt and deploy artificial intelligence. In its new State of Enterprise AI 2025 report, OpenAI describes how this massive user base has created a powerful consumer-to-enterprise flywheel, pushing AI deeper into global organizations and accelerating the shift from experimentation to operational scale.
OpenAI notes that general purpose technologies historically unlock the most economic value once businesses translate raw technical capabilities into scalable, repeatable workflows. According to the report, enterprise AI has now entered this phase. Organizations are embedding ChatGPT into key processes, deploying automated agents, and relying on multimodal AI systems to support decision-making, communication, and production.
The report draws on two major data sources: aggregated usage data from enterprise deployments of ChatGPT and a survey of 9,000 workers across nearly 100 global companies. Together, they offer one of the most detailed views yet of how AI is transforming modern work.
Enterprise adoption is rising sharply. Over the past year, weekly messages in ChatGPT Enterprise increased nearly 8 times, and the average employee is sending 30 percent more messages. Usage of structured workflows, such as Projects and Custom GPTs, rose 19 times year to date, showing a shift from casual querying to deeper, integrated workflows. Reasoning token consumption per organization climbed 320 times, revealing that businesses are increasingly using more advanced intelligence to power complex tasks.
Workers report that AI is not just accelerating existing tasks but enabling new ones that previously required specialized expertise. Time savings average 40 to 60 minutes per day, with heavy users reporting more than 10 hours weekly. Productivity gains span entire organizations, with 87 percent of IT teams resolving issues faster, 85 percent of marketing and product teams accelerating campaign execution, 75 percent of HR teams improving employee engagement, and 73 percent of engineers delivering code more quickly.
AI usage is expanding among nontechnical workers as well. Coding-related activity among employees outside technical roles increased 36 percent, narrowing the gap between intent and execution. Workers can now turn ideas into prototypes, analyses, or functional tools without requiring additional technical support.
Geographic adoption is accelerating too. The fastest-growing enterprise markets include Australia, Brazil, the Netherlands, and France, each exceeding 140 percent year-over-year growth. API adoption across international markets rose more than 70 percent in six months, with Japan now the largest API market outside the United States.
The report highlights a widening gap between frontier adopters and the median enterprise. Frontier workers send six times more messages than typical employees and make heavier use of advanced capabilities. Frontier firms send twice as many messages per seat and show deeper integration of AI across teams. According to OpenAI, productivity gains compound as organizations use more intelligence and apply AI across more tasks.
The biggest bottlenecks are no longer model performance or tool availability. Instead, organizations face constraints around readiness, training, governance, and implementation strategy. With OpenAI releasing new capabilities roughly every three days, companies must adapt quickly to keep pace with the shifting frontier.
The State of Enterprise AI 2025 report is intended to give executives and operators a clear benchmark for how leading organizations are deploying AI today, where they are realizing value, and how deeper integration amplifies impact over time. OpenAI says it will continue to expand its analysis as enterprises evolve from small-scale pilots to widespread, high-impact AI adoption.
