ChatGPT hits 900 million weekly users
OpenAI raises $110 billion as Amazon Nvidia and SoftBank bankroll scale, The free tier looks less like a product and more like a land grab
Images
Image Credits:Silas Stein/picture alliance / Getty Images
techcrunch.com
Aisha Malik
techcrunch.com
techcrunch.com
Aravind Srinivas, co-founder and chief executive officer of Perplexity, during TechCrunch Disrupt in San Francisco, California, US, on Wednesday, Oct. 30, 2024.
techcrunch.com
techcrunch.com
ChatGPT has reached 900 million weekly active users and 50 million paying subscribers, OpenAI said on Friday, as the company announced a $110 billion private funding round led by a $50 billion investment from Amazon. According to TechCrunch, OpenAI’s valuation in the deal is $730 billion pre-money, with Nvidia and SoftBank each committing $30 billion.
The numbers are a reminder that “AI” is no longer a research category so much as a distribution channel. At 900 million weekly users, the chatbot sits closer to the scale of the world’s largest consumer platforms than to enterprise software, and it is being financed accordingly: investors are effectively underwriting the cost of running the model for hundreds of millions of people until usage habits harden into defaults.
That subsidy is not abstract. Every extra prompt is inference compute that has to be paid for somewhere, and the funding round reads like a supply-chain agreement as much as a bet on growth. Business Insider reports that Amazon’s deal positions AWS as the exclusive third‑party cloud distributor for OpenAI’s Frontier business product and includes work on custom models for Amazon’s own applications. Nvidia’s participation is framed around expanding the compute available not just for training but for inference, the part of the stack that becomes more expensive as user numbers rise.
The corporate logic is straightforward: OpenAI’s backers are also Google’s competitors. Amazon competes with Google in cloud, advertising-adjacent product search, and custom AI chips; Microsoft has its own long-running rivalry with Google across cloud, productivity software and search. If ChatGPT becomes a default interface for “looking things up,” drafting documents, or routing tasks, it shifts traffic and bargaining power away from the incumbent gatekeepers.
For users and businesses, the same scale that makes the product feel ubiquitous also makes switching costs real. A chatbot that sits inside workflows—writing, planning, summarising, coding—quickly becomes a dependency. The bill can be deferred while venture capital and strategic investors absorb the compute burn, but it rarely disappears: it tends to reappear as higher subscription prices, tighter feature tiers, enterprise licensing, or deeper data collection to improve targeting and retention.
Governments usually arrive later, once the dependency is visible. A tool used weekly by close to a billion people is no longer just a private app; it is a piece of critical information infrastructure in practice, even if it is not treated that way in law. The next round of arguments—over safety standards, access, procurement rules, and who gets to set default behaviour—will not be settled by model benchmarks.
OpenAI’s own update was brief: 900 million weekly active users, 50 million subscribers, and a funding round still open to additional investors. The company did not say what it costs to serve that many users, only that the product is getting faster and more reliable as it scales.