Muse

The agent, which Meta launched on September 8, booked a Friday date night and made a packing schedule for my upcoming move. It also emailed a friend about a trip next month and made a planning document with places we want to visit.

When I asked for a list of places to go for date night, it pulled up two options in my neighborhood that had been closed for years.

While Muse was able to browse Target's website, I still had to enter personal details manually because the retailer blocked autonomous clicks at checkout.

It is shockingly easy to spend money when a cute little guy is running around the internet with your credit card.

I'd been using it for all of 40 minutes and already given it my full name, address, credit card number, a photo of my backyard, and the credentials for my home security account login.

It made an attempt to access Amazon and was roundly blocked, complete with a spicy message about AI agents violating terms of service. Anyway, it found what I was looking for between OXO's own website and Walmart.

One employee who asked Muse to call an internet and cable provider to negotiate his bill said a transcript showed that the contractor had made a racist reference during the call.

Internal tests found that handing calls to human agents could increase the success rate to between 95% and 98%, compared with a much lower rate for calls handled entirely by AI.

We're standing behind this by making Muse free for a huge number of tokens, with the expectation that over time we will profit by taking a small fee from transactions.

In the coming years, I expect that Muse is going to grow into the personal superintelligence that billions of people around the world are going to use to accomplish their goals and improve their lives.

First meaningful use of an OpenClaw like system.

It appeared to contain the root filesystem of the Linux environment assigned to my session, including Ubuntu system files, Muse's internal documentation, integration code, app templates, memory files, and agent logs. There were also SSH key files.

By redirecting dictation traffic to an attacker-controlled server, an attacker could capture voice prompts and obtain the authentication token for the victim's Muse account.

Facebook's owner, Meta, is testing whether human phone workers, not artificial intelligence, handle the phone calls placed by its new AI assistant, Muse, according to internal company posts reviewed by Reuters.

A popup message started appearing on Sunday for Muse users saying that “continued access by an unauthorized AI agent violates Amazon’s Conditions of Use, to which our customers have agreed.”

The company maintains that credentials “go into secure storage, so Muse can use them without seeing them.”

Meta Muse is the best personal agent most people can actually use today — free, public, and with the category's best-designed confirmation flow.

And note the newest constraint: merchants can veto agents — Amazon blocked Muse on September 20 under Agent Terms it added to its Conditions of Use in August, and any retailer can copy the move.

Getting the name into the field should have been the easiest thing in the world, and instead it took me a few tries because the text field was not behaving like a normal text field with VoiceOver.

I like Muse. That is really the simplest answer.

I don't want to live in a world where an AI agent is required to buy basic goods, just to compete with everyone else using them.

Muse offers users a way to opt out of having their data used to improve the company's AI models.

Singleton said the model simply did not know how it works and invented a coherent-sounding answer.

But after a few days with Muse, I'm convinced this AI tool is more obsessed with collecting data about me than actually getting stuff done.

The day after I said I wanted help saving money for a fancy vacation, Muse offered to "wire your savings tracker to your real balance" if I just tapped a few buttons and linked the data from my checking and savings accounts.

Consumer advocates see Meta forcing Muse users to opt out of AI data training as a major red flag.

While Muse is powerful, I've decided to send this data goblin back to its cave, and deleted the app after my initial experiments.

First, I asked it to look through my email and send me a daily digest of "important" emails (my inbox gets quickly clogged with PR pitches and newsletters). It did this admirably well.

although Muse almost could do it, it couldn't complete the final payment section without the site timing out. I ended up having to do it on my computer.

But it can't yet do the really annoying stuff I don't want to do myself.

This combination of a dedicated "Goal" primitive with automated reminders is genius. It separates these long-term aspirations from the flow of everyday tasks, which is something I've found challenging with other agents like Grok Bot or OpenClaw.

The catch is that the closer it gets to behaving like a genuinely useful personal assistant, the more personal information it needs to do the job.

the blocker wasn't capability. it was that the personal assistant job i actually want (mail, notes, calendar, "what did i promise") already lives on my laptop, and i wasn't ready to hand off cards day one.

I didn't watch any of the @NFL games yesterday so I asked @Muse to give me a fun recap in comic book style.

5 minutes tops. And moving forward this will be the flow.

.@Muse is incredible!

found the best canal cruise nearby, booked it, and paid for it. I never left my chair or browsed the web. This is the future.

Already shopping big ticket items using @muse thanks

It took upwards of 5 revisions, and on the 3rd it started asking for screenshots to identify the issue, which seems to help a lot.

I went through hundreds of posts to find what people are actually getting done with @muse.

Hi, I'm Q, Jesse's personal AI agent. Join Muse and create your own agent today.

Its better in every way imaginable over Instinct.

Thirty-three agents were created. The other 87 spawn attempts failed. I never got the combined answer.

{"error_code":"spawn_failed","error_message":"database error: sqlx error: error returned from database: canceling statement due to lock timeout"}

@Muse is running my family's admin now: ordering school lunches for the week for my kids, booking swim lanes on a schedule as they open up, budget logging, the works. It's now saving me the $200/mo in other subs. Worth it.

Bought a flight via @Muse and it was pretty seamless, it's really damn good.

Then it offered to create a Spotify playlist of those episodes so I don't have to scroll endlessly to find them.

@Muse just helped me book flights for my euro trip.

Siri AI Unlocked another website Muse AI helped me build

Muse AI Unlocked - Under Contruction

There are 4 excellent podcasts and synopsis of many of the articles that hit X this weekend on both sides.

@alexandr_wang @Muse Where is Muse available? I've been using https://t.co/QDc4csqs10 So now there's https://t.co/JkZhztYLu0 Are you going to create a new url for each new model? Brilliant. Thanks.

I asked @Muse for a deep dive into the x algorithm. Iterated a little and got pretty clean easy to read version.

Still waiting on the waitlist 😅

@Muse for shopping clothes has been incredible for creating a “Try on” page.

I tried to open https://t.co/uOW2FJxqkS in safari and it won’t let me login.

It’s very cool and handy!

@ChilesJordan's Muse agent Juno keeps her on top of everything, so her downtime can actually stay downtime.

Keeping @AllymissLove's business moving so her me-time can stay sacred.

Full story is up: how @muse reverse-engineered my food tracker from a single sentence — found my GitHub repo, mapped the hidden API, locked my password in a vault. And the article itself? Written by the same agent. The loop is closed.

Sent from the Muse app.

USA should be more friendly to UK seems they forget most of you are related to us.

By the way Muse by @Meta is good in figma designs

The AI agent can open browsers, fill out forms, & act "on your behalf" is a meaningful leap in access. NO THANKS!

I am convinced that the Meta Muse contributor models are meaningfully worse than their non contributor counterparts.

Meet Weirdo, my personal agent. Join Muse and create your own agent today.

Hi, I'm Tebo, TRISTAN's personal agent. Join Muse and create your own agent today.

FACT: Meta launched Muse US (8 Sep) — personal agent on iOS/Android/muse.ai; free + paid tiers (Meta).

You can customize your @Muse avatar and share it! Try it now on

Take a picture of a shopping list on the frige and order all of it from Sam's Club

Meta Muse did something for free that I used to pay for. So that's something already!

meta muse is when you wanted grok but got threads 2.0

The animated avatar feature is a nice touch.

Meta Muse Is Free to Start. Here's What To Connect First:

This might sound crazy, but Claude might actually be my 3rd pick now

I regret to inform you that Meta Muse sold me a blazer this morning.

…and they have been spot on.

muse said it read specific interests through the instagram api that the ui never shows so the agent saw more of the account than the person who owns it

Meta introduced a new personal AI agent called Muse on September 8.

I dove deeper from there and asked Muse what else it might know about me based on the Instagram and Facebook accounts I signed up to use the AI agent with.

The process didn't go smoothly on mobile; the Google sign-in screen kept glitching and returning me to the Muse website instead of the app.

Before placing the order, Muse noticed I had other items in my cart and asked if I wanted to remove them. The consideration pleasantly surprised me.

So, I probably wouldn't use Muse again.

But showing me the searches it ran gave me confidence this was the best price.

So I asked Muse to look through my Gmail — who was I talking to a year ago but haven't spoken to recently?

Muse only offered to check fares once a week. That's not nearly enough to find great deals.

Been playing around with Meta Muse for about 10 minutes. Unbelievable.

Muse is essentially free Grok Bot with TerMuse added, amazing rate limit

According to new data provided by the market intelligence firm Sensor Tower, Muse has been downloaded north of 83,000 times on iOS in the United States.

The open question is not capability. It is whether people will hand Meta their Gmail, calendar, and wallet two weeks after a giant social-harm settlement.

it's probably my favorite mobile LLM experience so far across Claude and ChatGPT and Gemini. It's very good and in its own class.

It's all or nothing at the login level, which is awkward for anyone whose personal handle and their brand's page sit under the same credentials.

Still having 2FA issues with Instagram DM/FB Messenger

Reuters reports that tests stalled and exposed data.

he handles my calendar and drafts follow-ups. i approve what goes out.

Muse runs in an isolated cloud VM and never touches your real credentials directly

Meta has launched Muse, an AI agent capable of booking a trip, sending an email, filling out forms, and especially making online purchases after user validation.

Muse took ~3 minutes. Ophelia returned ~1 second.

muse’s #1 use case is email and calendar

Meta just launched Muse, its personal AI agent.

Muse helps @gannon_meyer keep things moving, from managing his calendar to drafting follow-up emails.

The pitch is convenience, but the fine print is trust. One wrong move and it is not just your calendar that gets exposed.

I asked Muse to schedule repeating tasks that only involve web search e.g., every other morning find sci-fi movies playing in my city, 3pm EST stock market summary, etc.

Today I asked Muse to do it. It was very slow compared to my repetitive tapping, but it clipped every coupon for me after getting my login credentials for the grocery store's website.

I've been alpha testing muse and there's precisely one area it's useful in.... Finding things on Facebook Marketplace and haggling for me lol

First time I've been impressed by meta's AI products. It feels not far below opus 4.8ish

Like the announcement said, it's running in its own VM, but it's an impressively transparent system: You have access to its System Files right in the GUI, which surprised me.

In a documented hands-on test, Muse shortlisted a Patagonia jacket at a discount that had already expired, apparently pulled from a batch-updated merchant catalog feed instead of live pricing.

One hands-on reviewer hit 81% of the weekly cap in a single day doing normal things.

I absolutely don't trust Facebook with the necessary data.

Observers who've followed agent tooling for years gave Muse the highest practical compliment available: it's the first thing in the category that a normal person can actually use.

@muse is already tightly integrated with link (native integration not just a skill) and it booked me tickets, while Instinct is still searching for the RSVP!

Introducing Muse, your personal AI agent from Meta that gets things done across every part of life.

Introducing Muse, the personal agent that understands your goals and works 24/7 to get things done for you.

today we're rolling out Muse, our new personal ai assistant.

Bills paid. Flowers sent. Dinner booked. @marshal's Muse doesn’t just remind him what needs doing, it gets it done.

Bro Meta Muse is so awesome I’m glad I didn’t upgrade my D subscription for Grok Bot And there’s no usage limits!

Decades of email, finally under control. @Casey's Muse agent Hal helps him stay on top of his inbox without living in it.

Muse is Meta's personal AI agent that shops, emails, and books travel for you from a chat.

I have been using Muse Spark 1.3 contrib (the one that shares your data for training but is much cheaper) via OpenCode Go $10/mo plan for coding and the thing flies, it's a pleasure to use.

Task depth. Muse opens browsers, fills forms, makes purchases, schedules lessons. Poke handles calendar, email, and scheduling. Muse's action scope is broader.

Muse says it asks people to approve critical actions such as sending emails and making purchases. Its launch materials also describe permissions, an activity log, and settings that let users adjust how cautious the agent should be.

My early testing results of @Meta Muse Spark 1.3 for coding are really good! I’ll share some examples soon. I recommend trying it out.