OpenAI by the Numbers: What the Latest Stats Reveal About the Fastest-Scaling Software Company in History
The Numbers Are In — And They Are Extraordinary
OpenAI has been releasing data points throughout 2025 and into early 2026 that paint a picture of growth unlike anything the technology industry has seen before. Not faster than most companies. Faster than every company. The trajectory from a $3.5 million revenue startup in 2020 to a business generating over $1 billion per month by mid-2025 represents a 3,628x increase in five years — a pace that makes the early growth of Google, Facebook, and Netflix look gradual by comparison.
Here is what the latest numbers tell us, and more importantly, what they mean.
The Revenue Story
OpenAI generated $3.7 billion in revenue in 2024. By mid-2025, monthly revenue had crossed $1 billion — doubling from $500 million at the start of the year in just seven months. The company is now on track for $15-20 billion in annual revenue by the end of 2025, exceeding its own internal projection of $12.7 billion.
Annual recurring revenue hit $12 billion by July 2025, growing at 236% year-over-year. For context, it took Salesforce 18 years to reach $12 billion ARR. It took OpenAI roughly three.
The revenue breaks down across several segments. Consumer subscriptions — ChatGPT Plus at $20/month and ChatGPT Pro at $200/month — remain the largest contributor, commanding roughly 70% of the AI-as-a-service market. Enterprise revenue is growing even faster, with 600,000 paying business users representing 275% growth in twelve months. API and developer platform revenue accounts for approximately 15% of total revenue, with reasoning token usage up 320x year-over-year. Even education has become a meaningful segment at $230 million annually.
Perhaps the most telling metric: 92% of Fortune 500 companies now use OpenAI products.
The User Explosion
ChatGPT now reaches between 700 and 800 million weekly active users. Daily active users sit between 114 and 123 million. Monthly website visits exceed 5.4 billion. These are not niche technology numbers. They are global platform numbers — roughly 10% of the world's population interacting with ChatGPT every week.
The user base grew 25% between February and April 2025 alone, and has increased 8-10x since the 100 million weekly user milestone in November 2023. The demographics skew younger and male — 53-57% of users are between 18 and 34, and 64-72% are male — but the user base is broadening rapidly across age groups and geographies. The United States accounts for 14-17% of traffic, followed by India at 8%, Japan at 7.7%, and Brazil at 5.8%.
What are people doing with it? A study of 1.5 million conversations found that 49% of usage is asking questions, 40% is getting work done (writing and coding), and 11% is exploring ideas. Among developers specifically, 79% now rely on ChatGPT as part of their workflow.
The Money Behind the Growth
OpenAI's fundraising has been as unprecedented as its growth. The company has raised $57.9 billion across 11 rounds. The March 2025 round — $40 billion led by SoftBank — was the largest private technology fundraise in history. By early 2026, OpenAI entered talks for an additional $100 billion round at a $730 billion valuation, with NVIDIA contributing approximately $30 billion, Amazon over $20 billion, SoftBank another $30 billion, and Microsoft around $10 billion.
The Stargate Project, OpenAI's infrastructure initiative, represents $400-500 billion in planned spending for data centers with 7-10 gigawatts of capacity, creating an estimated 25,000 jobs across multiple US and international sites. Cloud commitments alone exceed $500 billion — $250 billion with Microsoft Azure, roughly $300 billion with Oracle over five years starting 2027, and $38 billion with AWS over seven years.
The Uncomfortable Truth: Losses Are Massive
For all the revenue growth, OpenAI is still losing significant money. The company reported losses between $9 and $14 billion in 2025, up from $5 billion in 2024. Cash burn reached $8.5 billion annually. Azure computing costs alone run $1.8 billion per year.
This is the fundamental tension at the heart of OpenAI's business. Revenue is growing at triple-digit percentages, but so are infrastructure costs. Every query, every API call, every token generated requires GPU compute that costs real money. The company projects reaching cash flow positive by 2029, with $2 billion in projected positive cash flow that year and a long-term revenue target of $100 billion.
Compute margins have improved to 70% by October 2025, which is encouraging. But the path from $15 billion in revenue with $14 billion in losses to sustainable profitability requires either dramatic cost reduction — likely through custom AI chips and more efficient models — or continued revenue growth that outpaces infrastructure spending.
What This Means for the AI Industry
AI is no longer experimental. When 92% of Fortune 500 companies are paying customers and 800 million people use a product weekly, the "will AI matter?" debate is over. The question now is how deeply it integrates into every workflow, every product, and every organization.
The market is concentrating — but not without competition. OpenAI holds 60%+ of the US AI-as-a-service market, but its enterprise foundation model share has dropped from 50% to 34% as Anthropic doubled its share to 24%. Google, Meta, and a growing ecosystem of open-source models are all competing aggressively. Market leadership is strong but not guaranteed.
Infrastructure is the real bottleneck. OpenAI has described itself as "constantly under compute." The hundreds of billions being invested in data centers and cloud commitments reflect a fundamental reality: the demand for AI compute is growing faster than the supply. This benefits chip makers like NVIDIA and infrastructure providers, and it creates opportunities for competitors who can offer more efficient architectures.
The economics are still unproven at scale. OpenAI's losses are eye-watering even by Silicon Valley standards. The bet is that AI will become so embedded in how people and businesses operate that the revenue will eventually dwarf the costs. History suggests this is possible — Amazon lost money for years before AWS became the most profitable cloud business in the world. But the scale of investment required here is unprecedented, and the competitive dynamics are more intense.
An IPO is on the horizon. OpenAI is reportedly targeting an H2 2026 filing with a 2027 listing, potentially at a $1 trillion valuation. If that happens, it would be one of the largest public offerings in history and a defining moment for the AI industry.
The Bottom Line
OpenAI's numbers tell a story of a company — and an industry — operating at a scale and pace that has no precedent in technology. Eight hundred million weekly users. A billion dollars in monthly revenue. Half a trillion dollars in infrastructure commitments. These are not projections or aspirations. They are current reality.
The question is no longer whether AI will transform how we work and live. The question is whether the economics can sustain the growth, whether the competition will fragment the market, and whether the infrastructure can be built fast enough to meet demand that shows no signs of slowing down.
Whatever happens next, the numbers make one thing clear: the AI revolution is not coming. It is already here, and it is moving faster than anyone predicted.