A billion users is a distribution fact

Gemini crossed a billion monthly actives. The number is real, the growth is real, and the denominator was chosen. All three things are worth holding at once.

Sundar Pichai announced a billion monthly active users for the Gemini app — the fastest-growing product in Google's 28-year history. ChatGPT passed a billion monthly in June and was reported at a billion weekly by July.

The curve is not in dispute

400 million monthly in May 2025. 650 million by October. 750 million in February 2026. 900 million at I/O in May. 950 million at the July earnings call. A billion now. That is steep, uninterrupted, and not the shape of a product being propped up.

The denominator was a decision

Google reported monthly actives against a competitor reporting weekly. A product with a billion weekly users necessarily has considerably more than a billion monthly ones, so the two announcements are not comparable and were never going to be. Choosing the metric on which you can say "a billion" is ordinary corporate communication, and it is worth noticing rather than being scandalised by.

Also absent: subscribers. Reach was announced. Conversion was not.

What the number actually proves

Distribution — which is the asset nobody can buy. Gemini reaches people through Android, Search, Workspace and Chrome with no acquisition step. Reaching a billion users is a different achievement when the users were already inside your products, and it is a durable one precisely because it is not purchased.

Set it against 87.5% of US venture dollars going to AI and the strategic picture sharpens. Startups are buying distribution at venture prices. Google already had it, and is now converting it at a rate no funding round can match.

The open question is the one Google did not answer: whether a billion people using something for free becomes a business, or stays a moat.

Google — Google's Gemini app hits 1 billion monthly active users → · TechCrunch — Google's Gemini app surges to one billion users → · Fortune — 87.5% of venture dollars went to AI →