About — Consumer AI Agent Tracker

What this is

If you're reading this because a newsletter landed in your inbox: this is a running record of how real businesses are actually putting AI chat/agents in front of their customers — not announcements or demos, but things a real customer could actually go use today. It's updated automatically, most days, by an AI research routine that searches the web, checks its findings against sources, and adds anything genuinely new to a permanent public record.

What counts as a "deployment"

To make the list, an AI agent has to be:

Consumer-facing. You, an ordinary customer, can talk to it — not an internal tool for employees, and not a tool a business uses to run its own operations.
Genuinely conversational. It has to handle open-ended, natural-language questions and hold a real back-and-forth — not a fixed menu of buttons or a keyword-triggered script pretending to be a chatbot. Plenty of "AI chatbots" out there are still the old scripted kind; those don't qualify.
Real and live. Launched and usable, with a dated, sourced reference — not a roadmap item or a press-release promise.

Business size doesn't matter — a five-person shop running an AI feature its booking software shipped counts just as much as a household-name retailer's own build. What matters is whether an actual customer can talk to it.

Two ways a business ends up on this list

Enterprise A large, well-known company builds and ships its own AI agent under its own name — think an airline's own booking assistant or a bank's own support chat.

Platform A smaller business doesn't build anything itself — it just switches on an AI feature that the software it already uses (its e-commerce platform, its booking system, its payments provider) ships. This is actually how most small and mid-sized businesses get an AI agent at all, so it's tracked by platform category rather than one company at a time — that way new entrants get caught automatically instead of the list going stale.

It doesn't stop at "it launched"

A launch announcement only tells you a company wanted good press that week — it doesn't tell you whether the thing is actually any good. So separately, this project follows up: hunting for hands-on reviews, comparisons, viral complaints, and plain evidence of whether an agent is genuinely useful or just a scripted FAQ bot wearing an AI label — and whether it knows when to hand off to a real person. That reception data lives alongside each entry once it's found, openly labeled by how trustworthy the source is (the company's own claim vs. independent reporting vs. real user complaints).

What you're reading right now

This newsletter is a digest — what's new since the last issue: freshly discovered deployments, and any new reception findings on ones already tracked. It's intentionally short. You can always find the most recent issue on the Latest News page, even if you're not subscribed. The full, permanently browsable, filterable record — every entry ever found, going back to the start — lives on the deployments page, with a companion platforms & providers page covering the AI vendors and tools referenced throughout.

Markets covered

United States
Canada
United Kingdom
France
Germany
Spain
Brazil
Mexico
Japan
Australia
India

Honesty about limits

This is best-effort, ongoing research, not an exhaustive registry — coverage is deeper in some markets than others, and a launch has to actually surface in a search to get caught. Every entry links back to its real source so you can check the work yourself rather than take it on faith. Nothing here is guessed or invented; where something is genuinely unclear or undisclosed, that's stated as an honest gap rather than filled in.