NEW Meta Muse AI Agent is ABSURD! 🤯

Vídeo em Inglês, publicado pelo canal Julian Goldie SEO.

Video transcript

Transcript from available captions; it may contain errors.

New Meta Muse AI agent is absurd. Today, I'm going to show you an AI that does not just talk. It actually goes and does the work for you. This thing plans your trip, it shops for you, it even sends your emails. It sounds fake, but it's real and it just dropped. Stick around because by the end of this you'll know if this AI is safe enough to trust with your whole digital life. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency Goldie Agency. While he's helping clients get more leads and customers, I'm here to help you get the latest AI updates and Julian Goldie reads every comment, so make sure you comment below. Today, we're talking about the new Meta Muse AI agent and honestly, the word absurd fits perfectly because some of what this thing can do sounds made up, but it's not. It's real and Meta just launched it. So, what is Muse? Muse is Meta's brand new personal AI agent. Most AI chat tools just answer your question and stop. Muse is different. You give it a goal and it makes a plan, uses your apps, browses the internet, fills out forms, and can even make purchases for you. Then here's the wild part. It can keep working even after you close the app. Let's make that simple. Old AI works like this. You ask a question, it gives an answer, done. Muse works like this instead. You give it a goal, it makes a plan, it takes action, it checks on things, then it finishes the job. That is a huge jump from a normal chatbot. Meta gave a real example of this. Someone saves a recipe on Instagram, Muse turns that recipe into a grocery list, then it builds a full dinner menu, then it remembers which guests can't eat certain foods, then it prepares invitations. One small idea turns into a full plan done by the AI. Now here's the part that made my jaw drop. You can close the app and Muse keeps working in the background. Picture this. You tell it, "Find me the best flight to Tokyo and book the best one." You say, "Okay, go do it." and you close the app. You go make lunch, you come back 30 minutes later and there's a message waiting. It says, "I found three flights. Here's the one I picked, ready to book." That is not a chatbot. That is a digital worker doing a real job for you while you're gone. And Muse also remembers things about you. Say you tell it once that your friend Sarah cannot eat dairy. Weeks later, you ask it to plan a dinner party. It remembers Sarah's rule without you saying it again. Most AI forgets everything the second the chat ends. Muse is built to actually hold on to details about your life and use them later. And it's not just planning trips and dinners. Meta says Muse can help with things like researching products and buying them for you, planning travel, and even helping negotiate a better deal. It can open up an actual browser, click around websites, and fill out forms just like a person would. You can literally watch it work in real time, and if you ever want to take over and finish something yourself, you can just jump back in. Now, let's get into how this thing actually works under the hood, because Meta didn't just slap a new feature onto an old chatbot. They built something much bigger. Meta gave Muse its own computer. It's called the Muse Secure VM. Think of it like this. Instead of Muse living inside your phone, it lives on its own private computer in the cloud. That computer has its own browser, its own storage, its own tools, and it's kept separate from everyone else's Muse. Meta's thinking here is simple. If an AI is going to act on its own, it needs its own space to work in, just like an employee needs their own desk and computer. If you're watching this thinking, "Okay, this is exactly the kind of thing that could save my business hours every single week." That's exactly why we built the AI Profit Boardroom. See, most business owners hear about a tool like Muse and think it sounds cool, but they never actually sit down and figure out how to use it in their own business. That's the gap we close. Inside the AI Profit Boardroom, we walk you through real AI automations step-by-step, so you're not just watching from the sidelines while everyone else figures it out first. We show you how to plug AI into the boring, repetitive parts of running a business, so you get more of your time back, and you're bringing in more leads and more customers without doing everything by hand. It's a free community built for people who want to actually use AI the smart way instead of just talking about it. So, if any part of what Muse can do got you thinking about your own business, come join us inside the AI Profit Boardroom, and let's build it out together. But, here's the obvious question. If an AI has its own computer and can act on its own, what stops it from doing something bad? That's where Sentinel comes in. Sentinel is like a security guard standing between Muse and the internet. Muse cannot just go do whatever it wants. Every single time Muse wants to take an action, like sending an email or buying something, it has to ask Sentinel first. Sentinel checks the request and either approves it or blocks it. Meta says Sentinel is the only thing that gets to say yes or no to Muse's actions online. So, Muse proposes the action, and Sentinel decides if it's allowed. For anything sensitive, like sending an email or making a payment, Muse has to stop and ask you for permission first, right there on your screen. So, Muse can plan, Muse can prepare, but you're still the one who says yes at the important moments. And here's another smart detail. When Muse is using its browser to click around a website, it's not just given full free run of the page. Meta says the browser agent only sees a simplified, controlled version of the page, and it's not allowed to run its own code on that page. That's another wall built in on purpose, so the AI can do its job without having too much power in the wrong place. Here's something else that's kind of wild. Muse isn't just limited to using the tools Meta gave it. It can build its own tools. It can create its own connections to other services. It can even run smaller helper versions of itself to handle different parts of a task at the same time. So, instead of one AI doing one thing, you've got one AI managing a whole team of smaller AI helpers underneath it. Let's talk about who can actually use this right now. Muse is live through the Muse app on the web and through WhatsApp, and it's coming to Meta's AI glasses down the line. Right now, it's only launched in the United States. So, if you're outside the US, like a lot of us watching this, you'll have to wait for it to roll out further. Meta says most people will be able to use it for free. And Meta already has their next move planned. Later this year, they want to release something called the confidential VM. The idea is your entire private AI computer gets locked with a key that only you hold. Meta is saying even they wouldn't be able to look inside your Muse's private computer. That's a big promise, and it's going to be interesting to see if they actually deliver on it. So, step back for a second and look at the bigger picture here. This isn't just one new feature. This is Meta saying we don't think AI should just answer questions anymore. We think AI should go out and do the work. That's a massive shift in how we're all going to use AI going forward. Now, here's the honest take. The upside is huge. Imagine handing off dozens of small annoying tasks to an AI that actually remembers what you like and gets them done while you're busy with something else. But, the more power we give an AI, the more that security, privacy, and permissions actually matter. So, don't treat this like magic. Treat it like a powerful new tool that still needs guardrails, which is exactly why Meta built Sentinel and all these safety layers in the first place. So, here's my question for you. Would you actually let an AI touch your email, your calendar, your shopping, and your browser so it also could handle your digital life for you? Drop your answer in the comments because Julian reads every single one, and I want to know if you're excited about this or if it freaks you out a little. If you want to actually put a tool like this to work growing your business, come join us inside the AI Profit Boardroom. It's a free community where we break down exactly how to use AI automation like Muse to save time, cut out the busy work, and bring in more leads and more customers. All explained in plain, simple steps so anyone can follow along. No tech background needed. We show you real examples every single week so you're not guessing what to do with AI, you're actually applying it in your own business right alongside us. And on top of that, if you want help making sure your content and your business actually get found on Google so all this AI automation work has somewhere to send traffic, book a free SEO strategy session with my team using the link in the the On that call, we'll look at where you're at right now and walk you through what it would actually take to get you more visibility, more traffic, and more customers finding you on search. It's completely free, there's no pressure, and it's a great next step if you're serious about growing your business with both AI and SEO working together. The chatbot era is ending, the agent era just started, and I'll see you in the next video.

Assistir ao vídeo original no YouTube

Assuntos: Agentes de IA, Muse. Publicado no YouTube em 2026-09-11.

Seleção automática por assunto e idioma. O vídeo pertence ao canal de origem.

About Muse checkout and Muse for Mac

When it is time to pay, Muse, the personal AI agent Meta launched on September 8, 2026, can check out with Link, built by Stripe. Meta says Muse is the first AI agent covered by Link's purchase protections, which include free coverage for damaged or lost items, price drops, no-fee returns and a return guarantee on eligible purchases. Link's wallet for agents generates a one-time-use card, so the real card details stay hidden. In the launch announcement, Meta listed Shop Pay as another payment option coming soon, together with 1Password support so Muse can use logins the person already has.

Muse's official page also describes Muse for Mac, which lets the agent work with files, apps and browser tabs on the user's computer, in addition to the mobile app and WhatsApp. According to the page's FAQ, Muse is free with a usage limit; when the limit is reached, the user can upgrade or wait for it to refresh. When Muse asks for permission, the options are allow once, always allow or deny, and the defaults can be managed for each connection.

Fact sheet

• Agent: Muse (Meta)
• Launch: September 8, 2026
• Checkout: Link, built by Stripe
• Card protection: One-time-use card number
• Purchase protections: Damaged or lost items, price drops, no-fee returns and return guarantee
• Announced as coming soon: Shop Pay and 1Password support
• Desktop app: Muse for Mac (files, apps and browser tabs)
• Free plan: Usage limit that refreshes, with upgrade option

Read also

• Meta Muse AI Is Scaring Me… It Knows WAY Too Much
• Master Meta's NEW Muse AI in 10 Minutes (What can it do?)
• Meta Muse AI Blew Me Away! Here’s How I’m Using It

---

Sources and references: Meta Newsroom – Introducing Muse · Meta AI – Muse · video by the Julian Goldie SEO channel on YouTube.