Meta promised an AI built from the ground up for privacy, then built a handy directory of strangers’ loan applications and bank records. Researchers needed one query. At this point Zuckerberg calling something »private« is less a security guarantee and more a spoiler warning.

As Ars Technica reported today, security researchers found Muse is capable of exposing sensitive information on its users through web links generated via a “share” feature.

It's always the share feature isn't it?

Not a particularly well-written article. The referenced Ars Technica article actually explains the problem:

Muse has allowed users to share searches, conversations, images, and similar content with other people. Such share features usually involve clicking an option to create a unique URL that anyone can use to load a page hosting the content. To keep potentially sensitive information from being accessible to unintended people, the URLs typically contain a long sequence of text that can’t be easily guessed.

... a central design decision for constructing URLs has made the links easy to discover. As a result, researchers from security firm Claroc said they were able to enumerate 6,800 links Muse users created for sharing conversations with others. After opening a few hundred of them, the researchers found information that was clearly not intended to be publicly available.

Claroc researchers were able to discover the Muse share links because all of them use the domain name cf.metaaiusercontent.com. Discovery is further made easy by the decision to add the conversation name to the URL, along with a long string of text... It’s this last design decision, putting the conversation title in the subdomain rather than in the URL path, that makes public discovery possible.

https://arstechnica.com/security/2026/10/meta-muse-exposes-users-personal-and-business-secrets-when-they-share-conversations/

TLDR;

They make dns records for each shared chat log. Dns records can then easily be looked up and accessed

Unsafe:
My-convo-with-PHI.aimetacontent.com
Safe:
Aimetacontent.com/l/random-hash/my-convo-with-PHI

investigators were able to easily open chat logs containing detailed private information, including corporate revenue documents, loan applications, bank transaction details, income details, and one workplace vacation calendar

I'm genuinely shocked that people thought it was a good idea to put any of that into Muse in the first place.

It would be a prudent move for your mental health to lower you're expectations a tiny bit

It's built for privacy in that its so bad that it causes a very small subset of people to start thinking about why privacy is a good thing actually...

"The engineer who spent last week writing it assured Zuck that every prompt used to write it closed with make no mistakes"

If that's not solid engineering then that engineer doesn't know what it is.

If you're trusting Meta with any information, you're a fool.

I would not call 9f21c8a3 long

Wow.

I believe Meta's founder and CEO himself uses the technical term "dumb fuck" to qualify these people.

I mean, they are airing ads about using their AI like a personal assistant, including to pay bills. So not just knowing financial details, literal approval to spend your money for you. I hope not many people are doing that, but I'm sure some are. Absolutely insane.