Urgent.News

What's breaking now, across thousands of outlets.

AI

Meta’s Muse AI Agent Tracks Down a User’s $1,500 in Unclaimed Funds

Meta’s Muse AI agent helped investment analyst Nick Grous find more than US$1,500 in unclaimed funds registered under his name after he asked it to search the internet for money that might belong to him. Grous, director of consumer internet and fintech research at ARK Invest, asked the agent to look online for money that […]

Meta’s Muse AI Agent Tracks Down a User’s $1,500 in Unclaimed Funds

Meta's Muse AI agent assisted investment analyst Nick Grous in locating over US$1,500 in unclaimed funds belonging to him, demonstrating the capabilities of AI agents today. Grous, a director of consumer internet and fintech research at ARK Invest, prompted the AI to scour the internet for any money that might be owed to him. Within a span of approximately five minutes, Muse generated a list of unclaimed property and funds associated with his name.

Grous later expressed his satisfaction, stating that he had become "a user for life." However, at the time of reporting, he had not yet received the funds. Muse located the funds and helped Grous complete most of the necessary paperwork, after which he filed an official claim. He mentioned that he would provide an update on whether the payment would be received.

This incident highlights one of the potential applications of AI agents: searching across websites, organizing information, and aiding users in completing forms. However, it also showcases the limitations of such automation. Despite this, the case illustrates the potential of AI agents like Muse, which can take actions on behalf of users once granted permission.

Muse can open web browsers, fill out forms, connect to emails and calendars, book travel, and perform other tasks. Meta claims that the agent can continue working on tasks even after the user closes the app. It operates within a secure virtual machine, isolating the agent and user's data from other systems. Users can interact with Muse via the app or through WhatsApp.

Initially launched in the United States, the service quickly gained millions of downloads, surpassing 3.4 million as of later estimates.

The episode sheds light on the workings of U.S. financial systems, where state governments hold substantial amounts of unclaimed property, including forgotten bank accounts, uncashed checks, deposits, and refunds. Users can search these official government databases, often for free, to identify potential matches and submit documentation to establish ownership.

This task is well-suited for AI agents due to their ability to automate repetitive searches and help organize information. However, it's important to note that AI agents cannot guarantee claim approval or payment receipt. Meta also promoted Muse through a social media campaign encouraging users to seek savings, refunds, or discounts.

While the company emphasized Muse's ability to perform tasks on behalf of users, questions about privacy and security have arisen due to its potential access to personal accounts and services. A cybersecurity consultant expressed skepticism about trusting Meta with such extensive control, while Amazon has blocked Muse from its website.

Meta has implemented security measures, including a dedicated virtual machine and a separate system controlling the agent's internet access. They also ensure that Muse cannot view users' passwords or payment information and seeks permission before performing sensitive actions like sending emails or making purchases. For Grous, the unclaimed-funds search exemplifies what AI agents can achieve with permission to navigate websites and handle paperwork. Whether the experiment ultimately leads to a payment remains uncertain.

Written by urgent.news from Colombia One's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at colombiaone.com →

More in AI

Trump says AI companies sign voluntary accord on safety controls

President Donald Trump and leaders of major artificial intelligence companies signed a voluntary accord Tuesday to strengthen safety reviews as the technology faces growing public concern. The agreement calls for internal controls, independent audits and board oversight, while Trump rejected calls for government restrictions and urged…

More from Tuesday 29 September →