By Katie Paul
NEW YORK, Sept 8 (Reuters) – Meta rolled out on Tuesday a long-touted AI assistant that can autonomously send emails, sell a car and book travel on a person’s behalf, despite internal concerns that the technology mismanages its access to sensitive personal data.
The company’s Muse agent, known internally as Hatch, is the centerpiece of CEO Mark Zuckerberg’s plan to offer “personal superintelligence” to the billions of people who use Meta’s services daily.
The product will be available only in the U.S. initially, via a dedicated Muse app or Meta’s WhatsApp messaging service, Meta said in its announcement. Meta said it plans to add the agent to its line of smart glasses “soon,” without elaborating.
Modeled on the open-source AI agent OpenClaw, Muse is designed to access a person’s apps across categories like email, calendar, payments, health, shopping and the smart home, Meta said. People choose which apps it connects to and can revoke access at any time.
Each Muse agent runs on its own virtual machine, a cloud-based emulation of a personal computer, which enables it to keep carrying out requests in the background even when a person is not actively using it.
Syncing up with apps containing a person’s real data increases the agent’s potential utility, while also significantly raising the stakes for safety and reliability issues, both for users who have entrusted it with their information and others who may be on the receiving end of agent misbehavior.
Vishal Shah, vice president of AI products at Meta, said the company had initially delayed the release of the product in April to make it more secure. Meta determined the extra work had allowed it to “cross the threshold” and meet its minimum requirements for product safety, security, privacy, model performance and other metrics.
“It is impossible to say that there is never going to be a mistake, but every single part of the architecture has been designed to make this as safe, as secure, as private as we can possibly make it,” Shah said in an interview.
INTERNAL TESTS REVEAL MIXED RESULTS
As recently as this week, Meta employees testing the tool have reported mixed results with Muse, with one person praising its utility in vacation planning and others describing cases in which it disconnected without explanation and uploaded sensitive information without permission, according to internal posts seen by Reuters.
One person wrote that the product had been so useful in arranging itineraries and ground transportation that it had become “the third participant” on a recent three-week honeymoon in Indonesia.
In another post, an employee who had prompted Muse to monitor for tickets and other items that sell out quickly reported encountering “many failure modes that made it unreliable.” The product stopped refreshing the page after about 15 minutes, silently ignored other errors and at times disabled monitoring “for no apparent reason,” the person said.
Meta Chief Technology Officer Andrew Bosworth posted that he kept getting logged out and needing to log back in, sometimes several times within a few minutes.
Others flagged serious security flaws, like an agent routing around guardrails to expose a person’s personal iCloud photos after being prompted to identify toys visible in pictures from a child’s birthday party.
Meta did not immediately respond to a request for comment on the specific incidents described in the internal posts.
(Reporting by Katie Paul in New York; Editing by Kenneth Li and Matthew Lewis)
Comments