Marc Zuckerberg explained how Muse AI works.
It can book reservations, make payments, and keeps your passwords and card details secure.
If the agent messes up a payment, it's insured so you'll get paid back.
It works via a chat interface in its own Muse app (and on WhatsApp).
---
Let's say you want to book a movie or pay for something:
1. @Stripe Link mints a single-use virtual card scoped to the purchase you approved. Muse never sees your real card.
2. If Muse makes a mistake, a refund guarantee covers it. XCover (Cover Genius) underwrites it, Link pays the premium.
3. Eligible purchases get price protection and damage or loss cover up to $500 per claim, plus no-fee returns up to $250.
Only 9% of consumers say they'd let an agent complete a purchase for them (Accenture, June 2026). I mean, if an agent buys the wrong flight, that sucks! The question holding agentic commerce back is who eats that loss.
An insured agent can be wrong and people will still let it spend. Insurance is the killer app for agentic commerce.
Card networks answered it in the 90s. People typed card numbers into websites because zero liability and chargebacks made a bad merchant the bank's problem. The difference there was the user was involved. The card networks will absolutely want to own this kind of protection. The industry needs to get ready for that.
On the business model, Zuckerberg told Alex Heath: Muse gives 100 million free tokens a week to win market share, $20 or $100 a month for power users, and eventually "a very small cut of whatever the transaction is."
So subscription plus payments.
There's a SURPRISING privacy-forward approach here:
He and Nat Friedman recruited Moxie Marlinspike, Signal's founder, to build a Confidential VM where "even Meta cannot see the content." That ships later this year.
And we already saw a privacy backlash with Instinct, which was storing your passwords and card credentials in its own vault. In the last two weeks it added Stripe Link and 1Password. Muse launched with both.
(You have to credit the partnerships folks at Stripe and 1Password here, locking in the personal agents super early.)
Where's Google in all of this personal agents discourse?
Google has the email, the calendar, the cards and the search history. Apple has a strong bit of user trust. Neither has shipped a personal agent.
This will become your control center and your interface. The everything app.
Is your consumer service ready for Muse as a customer? Is your Neobank? Your bank?
Is Meta a fintech now?
Is all AI fintech?
Mark Zuckerberg: the whole industry treats AI agents as a single-player game but the real unlock is agents interacting with each other, which Meta has been running internally
"Right now I think most of the industry is thinking about agents as like a single player game, right? It's like you have your agent and you use it."
"And there are going to be all these interesting things that basically you can do by having the agents interact with each other."
"And we already have all these interesting examples internally where people have their agents interacting with each other."
"This isn't, like, for the most part rolling out in this release, but it's going to be like an important part of how I think this works over time. As more of the people who you know start using Muse, it just gets better."
Dana White says Mark Zuckerberg is an animal and reveals Meta board meetings go from 9am to 6pm, dinner, then another three hours, extending into two days
"I'm on the board of Meta and I just got home last night from a Meta board meeting, and it's one of the best decisions I've ever made was when Mark Zuckerberg called me and I said yes to that."
"And you sit in this room, and not only the people that he runs through the room all day, which are some of their high-level employees there, but the rest of the board, you literally have the smartest people in the world trying to tackle problems that are, you know, a whole another level."
"There's levels to life and to business."
"So Zuckerberg is an animal, okay. These board meetings start at nine o'clock in the morning, they go till six, then you go to dinner and it's basically another three hours of the board meeting, right?"
"And every time we'll be in there, you know, two days in the room, and every time I walk out of those board meetings, I'm so fired up and inspired by the other board members and the people who work at Meta."
I don’t think people genuinely understand how insane this was
- First fight moving up to Heavyweight
- Bulked up 20-40+ lbs
- 3 years of inactivity from the octagon
Comes in there and beat their best HW in 3 minutes
G.O.A.T and it’s not even close
🚨HOLY SMOKES! 🇺🇸 PRESIDENT TRUMP IS CONTROLLING THE MIDDLE EAST! 👁️
🇸🇦 The Saudi Prince Bin Salman has just RE-ACTIVATED his OIL PIPELINE after shutting it down.
Saudi Arabia OFFICIALLY WITHDRAWS from the new Chinese BLOCKCHAIN system [mBridge].
This NEW Blockchain system has been created to DITCH the PETRO DOLLAR!
Pangram n’est pas fiable, mais, comme tout le monde répète qu’il l’est, il finit par le devenir aux yeux de tous. Je lui ai soumis un texte inédit de Balzac. Il s'agit de tout un paragraphe de « Le Lys dans la vallée » figurant dans le manuscrit, mais absent de l’édition originale comme des éditions suivantes. Il était donc impossible à Pangram d’aller retrouver ce passage sur Internet. Résultat de sa soi-disant analyse : 100 % IA. Il faut donc en conclure que Balzac a utilisé l’intelligence artificielle pour écrire ce chef-d’œuvre. Je viens d'envoyer, avec preuves à l'appui, un courriel aux concepteurs de Pangram.
🇫🇷🇨🇦🇭🇹📚🤖 FLASH | C’est la polémique qui anime le monde littéraire. L’écrivain Thélyson Orélien, qui vient de remporter le Prix du Roman Fnac 2026 pour « C’était ça ou mourir », phénomène de la rentrée littéraire, est accusé d’avoir écrit son livre presque entièrement avec l’IA.
Le compte X « Balance ton Claude » affirme avoir soumis de nombreux passages du roman au détecteur d’IA Pangram, qui en aurait identifié plusieurs comme étant générés à « 100 % » par une intelligence artificielle.
Des résultats qui alimentent les soupçons autour du premier roman de l’écrivain canado-haïtien, également sélectionné pour plusieurs grands prix littéraires. Son éditeur Grasset rejette ces accusations, tandis que la fiabilité des détecteurs d’IA pour déterminer l’origine d’un texte est également contestée par les défenseurs de l’auteur.
@JulienBroca@Monty_Brogan69 Beh justement non, c’est ce qu’il dit dans la vidéo. Il dit que c’est ce qu’il faut faire
Économie et investissement
T’as écouté Julien ou bien t’es juste intellectuellement hypocrite?
Le pire c’est que je suis pas d’accord avec lui, mais je fais preuve de sincérité, moi
@sacred_pop@ChristandGuitar I’ll play your game then.
You said that’s nowhere in the bible right?
Well your last sentence doesn’t appear in the bible neither
She sinned. Like Joseph did, like James did, like John the Baptist did, even tho all of their sins are not written in the bible
Checkmate, pagan
🚨🚨 Jérôme ROTHEN 🇫🇷 : « JE T’ANNONCE : SI MBAPPÉ GAGNE LE BALLON D’OR 🌕, déjà je pense que c’est un emplâtre (…) IL VA SE RETROUVER BIEN ESSEULÉ…
Vous vous rendez compte qu’il est en train d’acheter les voix au Ballon d’Or ? POUR AVOIR CE PUTA** DE BALLON D’OR ?… »
🎙️ @Rothensenflamme
Wild! Accused murderer requests physical demonstration at his own trial to prove he didn’t do it.
Then pauses and stares down the judge.
I've never seen anyone represent themselves so cleverly. Tomasz Kosowski is an evil genius. He may actually get away with it.