Instead of charts displaying project metrics for runs, users, tokens, and errors; I built these charts to display interesting datasets unrelated to the projects in this application.
ChatGPT pulled together datasets spanning 24 months (2023-09 → 2025-08) using public signals to estimate data for the big AI companies.
Because the companies don't publish clean month-by-month usage or revenue, normalized indices anchored to reputable datapoints are used instead of hard numbers to represent users and revenue figures.
OpenAI:Multiple reports cite ~$10B ARR (Jun 2025) and ~$12B+ by Jul 2025. I set July 2025 = 100 and backcast, so a value of 100 ≈ ~$12B ARR around 2025-07.
Anthropic:Reported $1B ARR (Dec 2024) → $3B run-rate by May 2025 → ~$4–5B by July 2025 (various outlets incl. Yahoo Finance/Bloomberg recap and The Information). A value of 100 ≈ ~$3B ARR around 2025-05 (later reports suggest ~$4-5B by July, thus >100 thereafter)
Gemini & xAI:Neither breaks out clean revenue and public figures are sparse. Proxy indices are used and derived from subscription/API signals and adoption momentum to give you a sensible chartable trend (not dollars).