Meta Muse Is Incredible - 5 Features You Need To Try

Summary
In this video, presenter Paul J. Lipsky provides a hands-on review and tutorial of Meta Muse, Meta’s personal AI consumer agent. He demonstrates five key features across desktop and mobile, including autonomous shopping and checkout, voice memo summaries, on-demand podcast generation, outbound business phone calls, and remote local file management, alongside Muse’s referral token tier.

What is shown

• Interface & Core Layout [00:20 - 02:42]: Lipsky tours the desktop app interface, demonstrating the single persistent chat model versus modular “Side chats,” proactive suggestions in the “Ideas” tab, and automated milestone tracking in the “Goals” tab.
• Autonomous Web Shopping & Secure Store [02:53 - 05:59]: Muse launches its built-in browser to search Facebook Marketplace and eBay for a used Sony ZV-E10 camera under $500, compares listings, prompts the user to log in via an encrypted “Secure Store” credential manager, and handles checkout approvals and Stripe Link payments.
• Voice Memos [07:18 - 08:25]: Lipsky requests a 1-minute audio memo summary of a long text discussion comparing Claude Fable 5.1 and GPT-6 Astra, showing how side chats can be configured to reply exclusively via audio notes.
• AI Podcast Generation & Automation [08:26 - 09:39]: Muse generates a 6-minute, 36-second dual-host podcast (“Astra vs Fable”) with playback speed controls, and Lipsky sets up an automated recurring morning research podcast covering daily consumer AI news.
• Autonomous Phone Calls [09:40 - 13:14]: Demonstrating early-access outbound calling to official US businesses, Muse generates a pre-call plan to check store stock at Barnes & Noble in Boulder, Colorado. Lipsky then tests a live call to his own business (“The Uncarved Block LLC”), where the AI caller (“Brett”) identifies itself as an AI assistant, asks for business hours, and produces a live transcript.
• Local Desktop File Access & Editing [13:24 - 15:54]: Running Muse on an always-on headless Mac Mini, Lipsky uses the mobile app to prompt Muse to read Intentional Open - Shortcuts Setup.md from the remote desktop, summarize it, and write the summary back into the file.
• Token Limits and Referrals [15:55 - 17:09]: Lipsky reviews the Usage settings menu showing a weekly free tier (15% consumed) and a referral program granting 1 billion tokens per invite up to 30 redemptions (30 billion tokens banked).

Claims & numbers

• The presenter claims Meta Muse is “without a doubt the easiest AI agent I’ve ever used.”
• The presenter claims phone calling is strictly limited to calling official US businesses for customer-side inquiries and cannot dial personal contacts or conduct cold sales outreach.
• The presenter states that when calling businesses, workers often recognize the caller is an AI and either hang up or react negatively.
• The presenter states that each referral earns both parties 1 billion Muse tokens, redeemable up to 30 times for a maximum of 30 billion extra tokens.
• The presenter claims competitor Grok Bot is “one or maybe five steps above what Muse can do” in terms of capabilities, though slightly more complex.

Notable quotes

• “Meta Muse is without a doubt the easiest AI agent I’ve ever used. But don’t let its simplicity fool you. Under the hood, it is incredibly powerful.” [00:00]
• “Anytime you ask it to create a phone call, it will always put together a plan first... Once you approve that, there’ll be another pop-up that asks you to approve it again... and then it will actually place the phone call.” [10:34]
• “I have gotten it to work and actually do tasks for me... but a lot of times, honestly, the businesses that I call get angry that an AI is calling, because they can tell.” [11:18]

Assessment
This is an authentic third-party tutorial and feature review showing real, unsimulated software interactions across the Meta Muse desktop and mobile apps. The presenter demonstrates functioning agent capabilities—including live phone calls, file read/write operations, and browser automation—while candidly discussing guardrails, limitations, and recipient reactions to AI phone calls.

Described by gemini-3.8-flash on 2026-10-13 from the video’s audio and frames.