ChatGPT: from a free research preview to $20 billion-plus in ARR
OpenAI launched ChatGPT as a free research preview on 30 November 2022. By the end of 2025 the company said annualized revenue had passed $20 billion, and in 2026 it reported more than 900 million weekly active users.
- Product
- ChatGPT
- Founder
- Sam Altman, Greg Brockman, and Ilya Sutskever
- Launched
- 2022-11
- Model
- Consumer and business subscriptions, usage-based APIs, and a free tier that the company has also used for ads and commerce
Published · 3 min read · 8 sources
Key takeaways
- OpenAI introduced ChatGPT on 30 November 2022 as a free research preview, fine-tuned from a GPT-3.5 model.
- CFO Sarah Friar wrote in January 2026 that revenue grew from $2 billion ARR in 2023 to $6 billion in 2024 and $20 billion-plus in 2025 (company-reported).
- In March 2026 OpenAI said it had closed $122 billion in funding at an $852 billion post-money valuation, and that ChatGPT had more than 900 million weekly active users and over 50 million subscribers (company-reported).
What it is
ChatGPT is a conversational AI assistant from OpenAI. On 30 November 2022 the company introduced it as a research preview "to get users' feedback and learn about its strengths and weaknesses," and said usage was free during that preview (OpenAI). OpenAI itself was announced on 11 December 2015 as a non-profit research company whose co-chairs were Sam Altman and Elon Musk, with Greg Brockman as CTO and Ilya Sutskever as research director (OpenAI).
Timeline
- December 11, 2015: OpenAI launched as a non-profit, saying funders including Altman, Brockman, Musk, Reid Hoffman, Jessica Livingston, Peter Thiel, AWS, Infosys, and YC Research had committed $1 billion, "although we expect to only spend a tiny fraction of this in the next few years" (OpenAI).
- November 30, 2022: OpenAI published "Introducing ChatGPT," describing a GPT-3.5 model trained with reinforcement learning from human feedback, and invited people to try it at chatgpt.com (OpenAI).
- October 2, 2024: The company said it had raised $6.6 billion at a $157 billion post-money valuation, and that "over 250 million people" used ChatGPT every week (company-reported) (OpenAI).
- March 31, 2025: OpenAI announced $40 billion in new funding at a $300 billion post-money valuation, and said 500 million people used ChatGPT every week (company-reported) (OpenAI).
- January 18, 2026: CFO Sarah Friar wrote that revenue had grown "3X year over year, or 10X from 2023 to 2025: $2B ARR in 2023, $6B in 2024, and $20B+ in 2025" (company-reported) (OpenAI).
- February 27, 2026: OpenAI announced $110 billion in new investment at a $730 billion pre-money valuation, including $30 billion each from SoftBank and NVIDIA and $50 billion from Amazon. It said ChatGPT had "more than 900M weekly active users" and "more than 50 million consumer subscribers" (company-reported) (OpenAI).
- March 31, 2026: The company said it had closed $122 billion in committed capital at an $852 billion post-money valuation. It also wrote that it was "now generating $2B in revenue per month," that enterprise "now makes up more than 40% of our revenue," and that an ads pilot "reached more than $100 million in ARR in under six weeks" (all company-reported) (OpenAI).
Why it worked
- It shipped a usable product, then monetized demand. Friar wrote that OpenAI launched ChatGPT as a research preview "to understand what would happen if we put frontier intelligence directly in people's hands," then added consumer subscriptions, workplace plans, usage-based API pricing, and later commerce and advertising (OpenAI).
- Distribution came first. The March 2026 post argued that "the broad consumer reach of ChatGPT creates a powerful distribution channel into the workplace" (OpenAI).
- The company raised against a very expensive compute constraint. Friar wrote that available compute grew from 0.2 GW in 2023 to about 1.9 GW in 2025, and that "revenue followed the same curve" (OpenAI).
What to learn
- A free preview can be a go-to-market, not just a demo. ChatGPT launched with no price and an invitation to try it (OpenAI).
- Company-reported ARR and weekly users can move quickly; treat each figure as a snapshot from a specific post, not an audited series (OpenAI).
- A consumer front door can subsidize an enterprise motion. OpenAI later said workplace demand was "rapidly shifting from basic model access to intelligent systems" and that enterprise was already more than 40% of revenue (OpenAI).
- Monetization can stack. Friar described subscriptions, a free ad- and commerce-supported tier, and usage-based APIs as a "multi-tier system" (OpenAI).
Risks and what didn't work
- The 2015 structure did not last. OpenAI launched as a non-profit "unconstrained by a need to generate financial return" (OpenAI). The later funding posts describe a business raising at very large valuations; this article does not reconstruct the later corporate restructuring from those pages alone.
- Revenue, user, and subscriber counts are company-reported and are not independently audited in the posts cited (OpenAI; OpenAI).
- Compute commitments are large and lumpy. Friar wrote that "growth does not move in a perfectly smooth line" and that "at times, capacity leads usage" (OpenAI).
- Ads and commerce are a newer, less proven layer. The March 2026 post said an ads pilot reached more than $100 million ARR in under six weeks (company-reported); that is a run-rate claim from a pilot, not a full-year result (OpenAI).
FAQ
Short answers drawn from the sections above.
What is ChatGPT?
ChatGPT is a conversational AI assistant from OpenAI. On 30 November 2022 the company introduced it as a research preview "to get users' feedback and learn about its strengths and weaknesses," and said usage was free during that preview (OpenAI). OpenAI itself was announced on 11 December 2015 as a non-profit research company whose co-chairs were Sam Altman and Elon Musk, with Greg Brockman as CTO and Ilya Sutskever as research director (OpenAI).
Why did ChatGPT work?
- It shipped a usable product, then monetized demand. Friar wrote that OpenAI launched ChatGPT as a research preview "to understand what would happen if we put frontier intelligence directly in people's hands," then added consumer subscriptions, workplace plans, usage-based API pricing, and later commerce and advertising (OpenAI).
- Distribution came first. The March 2026 post argued that "the broad consumer reach of ChatGPT creates a powerful distribution channel into the workplace" (OpenAI).
- The company raised against a very expensive compute constraint. Friar wrote that available compute grew from 0.2 GW in 2023 to about 1.9 GW in 2025, and that "revenue followed the same curve" (OpenAI).
What can indie builders learn from ChatGPT?
- A free preview can be a go-to-market, not just a demo. ChatGPT launched with no price and an invitation to try it (OpenAI).
- Company-reported ARR and weekly users can move quickly; treat each figure as a snapshot from a specific post, not an audited series (OpenAI).
- A consumer front door can subsidize an enterprise motion. OpenAI later said workplace demand was "rapidly shifting from basic model access to intelligent systems" and that enterprise was already more than 40% of revenue (OpenAI).
- Monetization can stack. Friar described subscriptions, a free ad- and commerce-supported tier, and usage-based APIs as a "multi-tier system" (OpenAI).
What didn't work for ChatGPT, and what are the risks?
- The 2015 structure did not last. OpenAI launched as a non-profit "unconstrained by a need to generate financial return" (OpenAI). The later funding posts describe a business raising at very large valuations; this article does not reconstruct the later corporate restructuring from those pages alone.
- Revenue, user, and subscriber counts are company-reported and are not independently audited in the posts cited (OpenAI; OpenAI).
- Compute commitments are large and lumpy. Friar wrote that "growth does not move in a perfectly smooth line" and that "at times, capacity leads usage" (OpenAI).
- Ads and commerce are a newer, less proven layer. The March 2026 post said an ads pilot reached more than $100 million ARR in under six weeks (company-reported); that is a run-rate claim from a pilot, not a full-year result (OpenAI).
Sources
Every link cited above, in order of first use.
- 1.ChatGPT (chatgpt.com)
- 2.OpenAI (openai.com)
- 3.OpenAI (openai.com)
- 4.$6.6 billion (openai.com)
- 5.$40 billion (openai.com)
- 6.$2B (openai.com)
- 7.$110 billion (openai.com)
- 8.$122 billion (openai.com)
Related case studies
More in AI tools →Shopify: from a 2004 snowboard shop to $11.6 billion in 2025 revenue
Tobi Lütke built Shopify after the software for his own snowboard store proved more interesting than the shop. The company launched in 2006. In a May 2015 letter to shareholders Lütke cited more than 165,000 stores; for 2025 Shopify reported $11.6 billion in revenue.
Claude: from a March 2023 launch to a $47 billion revenue run rate
Anthropic, a public benefit corporation founded by Dario and Daniela Amodei, launched Claude in March 2023. In May 2026 the company said it had raised a $65 billion Series H at a $965 billion valuation, with a company-reported revenue run rate above $47 billion.
Devin: Cognition's AI software engineer from $1M to $73M ARR in nine months
Scott Wu's Cognition launched Devin in 2024. Company posts say Devin ARR grew from $1M (Sep 2024) to $73M (Jun 2025), then Cognition raised at $10.2B and later $48B valuations.