The data · sourced & dated

Chosen by some.
Shipped to everyone.

Agentic AI arrives through two doors: people choosing it, and platforms rolling it out by default. Here's how fast it's moving — how reliable it is today — and where it's heading.

How far it's spread

The numbers, on the record

Door one · by choice
0M
People choosing to use ChatGPT every week
Door two · by default
0M+
GenAI-capable smartphones already shipped — the agent arrives in an update
What it adds up to
+0%
Year-over-year growth in AI-referred traffic to US retail sites, Q1 2026

FIGURES LAST CHECKED 5 JULY 2026 · EACH LINKS TO ITS SOURCE — IF A NUMBER HERE IS STALE, TELL US.

No install step

It's already built in

Default, not download

On the device you already own

Voice assistants and AI features now ship by default in iOS and Android, in major browsers, and in the productivity tools people use at work. Adoption isn't a choice users make — it arrives in an update.

Honest caveat

How reliable is it — today

Right now, agents still mis-read, mis-book, and hallucinate. Reliability is climbing fast but uneven. We'll keep this page dated, because "currently" is a moving target — don't trust an undated claim about AI reliability.

The horizon

Limits today, direction tomorrow

Today's limits

Context & language

Finite context windows, uneven coverage across languages and regions, and stale training cutoffs all cap what an agent reliably knows right now.

Heading

What users push for

People want agents that just handle it — fewer prompts, more follow-through. That pressure pulls the whole field toward autonomy.

Heading

Better parameters

More capable defaults and "auto" modes (e.g. Claude picking its own depth) push agents from assistant toward something that acts on its own.

The curve won't wait for you

By the time agentic discovery is the default, the shelf will already be set. The businesses legible now are the ones being read.

Get on the shelf →