Meta's Muse AI Agent Launches to Mixed Reception, Hits No. 2 App Spot

An AI that doesn't just inform you but executes tasks in the real world
Muse represents a shift from AI assistants that answer questions to agents that spend your money and make commitments on your behalf.
Mark

So Muse hit No. 2 in the app store basically overnight. What's driving that kind of speed?

Mimi

Travel planning is genuinely annoying for a lot of people. Muse does the whole thing—search, compare, book—and handles payment. You tell it what you want and it handles the rest.

Luke

But do we know how many of those downloads are people actually using it versus people trying it once? App store rankings don't tell you retention.

Mark

Fair point. What about the pricing? Three tiers—do we know what separates them?

Mimi

The reporting doesn't get into the specifics of what each tier includes. Just that there are three levels and they integrate with payment systems.

Luke

That's a gap. You can't really evaluate whether this is a sustainable business without knowing what you're actually paying for at each level.

Mark

And the creepy factor—the reviews mentioned that. What's the actual concern?

Mimi

An AI agent that has your payment information and can spend your money, even within limits you set. Some people find that uncomfortable, even if it works.

Luke

Which is a legitimate concern, but we should separate two things: does it work reliably, and are people comfortable with the premise? Those are different questions.

Mark

So it might work perfectly and still fail because people don't want to use it?

Mimi

Exactly. Or it might work perfectly and succeed because convenience wins out over discomfort. We won't know for a few months.

Luke

The real test is retention. Anyone can download an app. The question is whether people are still using Muse in October.

  • Muse arrived and immediately disrupted app store rankings, reaching No. 2 in the US within days — a velocity that signals something more than novelty.
  • The app's ability to hold payment information and execute real-world transactions on a user's behalf has unsettled privacy advocates and cautious consumers alike.
  • Meta is navigating the tension between mass adoption and sustained trust through a three-tier pricing model designed to convert curious downloaders into committed, paying users.
  • Early reports suggest the travel booking functionality works reliably, but reliability and genuine comfort with AI autonomy are proving to be very different thresholds for users to cross.
  • The real test is retention — whether Muse becomes a habitual tool or a novelty people abandon after a single use will determine whether this launch is a turning point or a footnote.

Meta's release of Muse marks a quiet but consequential moment in the long negotiation between human autonomy and technological convenience. Within days of its launch, the personal AI agent — capable of booking flights, reserving hotels, and completing payments on a user's behalf — became the second-most downloaded app in the United States. The speed of its adoption suggests that a threshold has been crossed: enough people are now willing to hand a machine not just their questions, but their money and their decisions. Whether that willingness reflects trust earned or caution abandoned remains the deeper question.

Meta released Muse this week, and within days it had climbed to the second-most downloaded app in the United States. The product is a personal AI agent built to handle travel logistics — flights, hotels, reservations — and to complete those transactions directly, without requiring the user to visit individual websites or repeatedly enter payment details. Three pricing tiers suggest Meta is betting on long-term engagement, not just launch curiosity.

The speed of adoption is the most striking part of the story. App store rankings are shaped by millions of individual decisions, and they move slowly — except when they don't. Muse's rapid ascent suggests that something has shifted in how ordinary people think about AI assistants. A year ago, handing an AI your payment credentials and letting it book your vacation would have seemed reckless to most. Now it appears to be mainstream enough to drive one of the fastest climbs in recent app store history.

Reception has been genuinely mixed. Some users find the travel booking functionality useful and reliable; early reports suggest it performs well for the tasks it's designed for. Others remain uneasy with the premise itself — an AI that doesn't just answer questions but acts in the world, spending your money and making commitments on your behalf. Reliability and comfort, it turns out, are not the same thing.

The broader question is whether adoption holds. Launch momentum can carry an app to the top of the charts and then evaporate. What Meta is really testing is whether enough people find Muse useful enough to keep paying for it — and whether the appetite for AI agents that execute real-world tasks is as durable as this week's numbers suggest.

Meta released Muse this week, and within days it became the second-most downloaded app in the United States. The product is a personal AI agent designed to handle tasks that typically require human attention—booking flights, arranging hotels, managing reservations, and handling other day-to-day logistics. It arrives with three pricing tiers and the ability to process payments directly, meaning the AI can not only plan your trip but actually complete the transaction on your behalf.

The speed of adoption has been striking. App store rankings move slowly, shaped by millions of individual download decisions, yet Muse climbed to the No. 2 position almost immediately. That kind of velocity suggests something has shifted in how people think about AI assistants. A year ago, the idea of handing an AI agent your payment information and letting it book your vacation would have seemed reckless to most users. Now it appears enough people are willing to try it that the app has become a mainstream phenomenon.

What Muse does, technically, is straightforward. You tell it what you need—a weekend in Portland, a hotel near the waterfront, a flight that leaves in the morning. The agent handles the search, comparison, and booking across multiple platforms. It integrates with payment systems so the entire transaction happens without you having to visit individual websites or enter your card details repeatedly. For people who find travel planning tedious or overwhelming, this is genuinely useful. For people who like control over their bookings, it's less clear what the appeal is.

Meta's three-tier pricing structure suggests the company is betting on sustained engagement. The tiers presumably offer different levels of capability or usage limits, though the specifics of what each tier includes remain somewhat opaque in the available reporting. What's clear is that Meta sees Muse not as a one-time tool but as something people will pay for repeatedly, which means they're betting the agent will become a regular part of how users manage their lives.

The reception has been mixed, which is worth noting. Some reviewers have expressed discomfort with the premise itself—the idea of an AI agent having access to your payment information and the autonomy to spend your money, even within parameters you set. Others have focused on the practical question of whether the agent actually works reliably. Early reports suggest it does, at least for the travel booking use case it's designed for. But reliability and comfort are different things. You can trust a system to work correctly and still feel uneasy about what it represents.

What matters now is whether this adoption holds. App store rankings are volatile. A product can hit No. 2 on launch momentum and novelty alone, then fade quickly once the early adopters have tried it. The question is whether Muse becomes something people actually use regularly, or whether it becomes a curiosity people download, test once, and abandon. The three-tier pricing model suggests Meta is prepared for some churn—they're not betting everyone who downloads will stay, just that enough will to make the business work.

The broader signal here is about consumer appetite for AI agents that can act on their behalf. For years, the AI conversation has centered on chatbots that answer questions or generate text. Muse represents a different category: an AI that doesn't just inform you but executes tasks in the real world, spending your money and making commitments on your behalf. If Muse succeeds at scale, it won't be because the technology is revolutionary. It will be because enough people decided that the convenience of having an AI handle their travel logistics outweighs their concerns about privacy, control, and what it means to delegate that much authority to a machine.

Meta's Muse works and creeps me out
— The Verge reviewer, capturing the tension between functionality and discomfort with AI autonomy
Envie de l'histoire complète ? Lire l'original sur Google News ↗
Nous contacter FAQ