Meta wants Muse to answer your email, book your holiday and pay for things on your behalf. If you live in Amsterdam, you can read about it, but you can’t install it. Welcome to the agent era, Dutch edition.
What Muse does
Muse is an agent, not a chatbot. You give it a task, and it works through it: it browses the web, makes purchases, generates images, drafts documents, and connects to the apps you allow through connectors. It can also set reminders, track goals and keep an eye on things you care about. Tasks keep running in the background, even after you close the app.
That’s the difference that matters. A chatbot answers. An agent does the errand.
Where it works
Muse launched in the United States in September 2026. The Muse app is available in the US and Canada, and the iPad version is limited to the same countries. Meta has not announced a date for Europe.
Press coverage gives reasons: privacy rules and regulatory friction. Meta hasn’t confirmed them. Whatever the cause, the practical answer is simple. If you’re in the Netherlands, Muse is not an option today.
The good points
- Separate watchdog. A second agent, Sentinel, monitors what Muse does and tries to block malicious instructions hidden in web pages.
- Isolated environment. Muse runs in its own virtual machine, so it doesn’t work directly on your device.
- Passwords and card details stay hidden. According to Meta, Muse can’t see them. Payments run through Stripe Link with one-time card numbers.
- You choose the access. You decide which apps Muse can reach, and you can disconnect them at any time.
- Ad-free promise. Meta says the agent’s conversations and data stay out of its advertising system.
These are real design choices. Most agents don’t offer that level of separation.
The bad points
- Trust is the catch. An agent that reads your email, handles your money and sees your calendar is a valuable target. The safeguards are only as good as their track record, and Muse is too new to have one.
- Promises are not proof. “Stays out of advertising” is a statement by the company that sells advertising. Until an independent audit shows otherwise, treat it as a claim.
- Adults only, US only. Those limits are at launch. They say a lot about how much Meta expects to get wrong in the first months.
- It’s Meta. The company’s record on personal data is the reason many Europeans are cautious about this product. Regulators have the same concern.
What this means for you
Don’t build a workflow around Muse yet. If you want an agent for email or admin today, check what’s actually available in the EU, read the permissions carefully, and give the agent access only to the apps it needs.
If you’re a developer, the more practical route is Muse Wonder through Meta’s developer API. Check your local availability before you plan around it.
Conclusion
Muse is a real step from chatbot to agent, and its security design is more thoughtful than most. But right now it’s a US product, and the trust question is still open. Watch the Europe announcement, and when it comes, read the privacy terms before you click Accept.
