synthetic.work · Original source page
Image provenance
Inherited source visual. Image capture date and exact event relationship were not established again in this expansion. Owner publication review pending; credit does not grant permission.
Original assetThe signal
On 1 February 2023, Reuters reported that ChatGPT was estimated to have reached '100 million monthly active users' in January, two months after launch, citing a UBS study. Signal Atlas's AI sector has few adoption claims as widely repeated as this one, yet the number did not originate with OpenAI. OpenAI's own launch post, dated 30 November 2022, introduces the model and its research-preview terms and states no user count at all.
The evidence
The Reuters report says UBS's estimate drew on data from the analytics firm Similarweb, which found roughly '13 million unique visitors' used ChatGPT per day in January, more than double December's level, and that Similarweb did not confirm the figures on request. The article also cites Sensor Tower data to note that reaching 100 million users took TikTok about nine months and Instagram two and a half years, framing ChatGPT's pace as unusually fast against that comparison set. None of this is a company disclosure: it is a bank's estimate, built on third-party web-traffic panel data, reported by a news organisation. Seven months later, OpenAI's own Enterprise announcement, dated 28 August 2023, does not repeat a user count; instead it states that 'teams adopt it in over 80% of Fortune 500 companies,' a different metric entirely.
Timeframe and confidence
The underlying traffic estimate covers a single month, January 2023, from a panel-based analytics method that estimates rather than directly counts every user. This is an editorial reading, not a criticism of the panel method as such: web-analytics estimates carry sampling and extrapolation error that a company's own logged-in-user count would not. The claim is best read as evidence that growth was fast by the comparators cited, not as a precise census of ChatGPT's user base in January 2023.
What would change the reading
A later OpenAI disclosure of a comparable, precisely defined user count for the same period would let the estimate be checked against a company figure. Its absence in the sources examined here means the estimate stands unverified by the company itself.
- Did OpenAI ever publish a monthly-active-user figure with a stated definition for early 2023?
- What does 'unique visitor' mean in Similarweb's method, and does it match what a company would call an active user?
- When a fast-adoption claim is repeated widely, has anyone checked the original estimate's method, or just its headline number?
The claim spread because it was striking and because two named comparators made it legible. Its evidentiary base is a bank's analysis of third-party traffic data, not a filing or a company statement, and that distinction is the more durable finding here.
Source trail
- Introducing ChatGPTopenai.com · Source publication: 2022-11-30 · Retrieved 2026-09-16
OpenAI's own launch post, which names ChatGPT's release date and describes the model without stating any user count.
- ChatGPT sets record for fastest-growing user base - analyst notewww.reuters.com · Source publication: 2023-02-01 · Retrieved 2026-09-16
Documents the UBS analyst estimate of 100 million January users, its Similarweb data basis, and the TikTok/Instagram comparators.
- Introducing ChatGPT Enterpriseopenai.com · Source publication: 2023-08-28 · Retrieved 2026-09-16
Shows OpenAI's later adoption framing (Fortune 500 penetration) without restating or confirming a user-count figure.
- Event date
- 2023-02-01
- First source date
- 2022-11-30
- Source-record publication
- Not supplied — draft retained
- Preparation
- 2026-09-16