8亿周活、Token消耗量激增320倍……OpenAI最新报告：企业AI已进入“深水区”

Core Insights - OpenAI's report indicates that AI is penetrating various industries, with technology, healthcare, and manufacturing being the fastest-growing sectors [1] - The report reveals that ChatGPT's weekly active user base has surpassed 800 million [1] - The application of AI in enterprises is characterized by accelerated penetration and deep integration, with significant increases in usage metrics [1] Group 1: AI Adoption and Usage Metrics - The average weekly interaction volume for ChatGPT's enterprise version has increased approximately 8 times over the past year, with individual users sending 30% more commands weekly [1] - The frequency of structured workflows, such as project management and customized GPT applications, has surged 19 times this year, indicating a shift from ad-hoc queries to systematic business processes [1] - The average token consumption for model inference per enterprise has grown about 320 times in the last 12 months, reflecting the systematic integration of advanced models into expanding product and service frameworks [1] Group 2: Industry and Regional Growth - AI is being utilized for increasingly complex professional tasks, with strong growth observed in professional services, finance, and technology sectors [2] - In the global market, enterprise customer numbers in Australia, Brazil, the Netherlands, and France have increased by over 140% year-on-year, marking them as the fastest-growing regional markets [2] - OpenAI's international API enterprise customer base has grown by over 70% in the past six months, with Japan having the highest number of enterprise API customers outside the U.S. [2] Group 3: Efficiency Gains and User Trends - 75% of surveyed employees reported significant improvements in work output efficiency or quality after using AI tools, saving an average of 40 to 60 minutes per day [2] - High-frequency users save over 10 hours of work per week, and the volume of programming-related commands from non-technical staff has increased by 36% [2] - 75% of users have successfully completed tasks that they could not have done independently before using AI [2] Group 4: Disparities Among Users and Organizations - A widening gap exists between leading enterprises and the industry average, with top users sending commands six times more than average users and leading companies achieving double the weekly interaction volume of the industry median [3] - The report emphasizes that the core bottleneck for enterprises adopting AI is no longer model performance but rather organizational adaptability and execution capabilities [3] - OpenAI is releasing new features or capabilities approximately every three days, highlighting the rapid pace of innovation in AI tools [3]