The Daily PulseTechnology
Meta's New Muse AI Agent Surges in Popularity While Facing Amazon Block and Security Flaws
Reported by Siddharth Deshmukh (Staff Writer) · Indian Express - Technology ·
Confirmed by a second source: TechCrunch
Representative image · PexelsMeta's personal AI agent app, Muse, has quickly climbed to the top of mobile app charts since its launch on September 8. In its first six days, the app recorded over 902,000 downloads on Apple's App Store and Google Play Store, making it the number one free app in the US and trailing only ChatGPT in post-launch download numbers. Powered by the Muse Spark large language model, the macOS application is designed to help users handle tasks like filling out electronic forms, organizing email inboxes, booking travel, turning recipes into grocery lists, and making purchases.
Amazon Block and Revenue
The app achieves these tasks by using built-in connectors to navigate a user's emails, calendar, payment apps, health tools, smart home devices, and social media accounts. This deep access has already caused friction with online service providers. Amazon blocked Muse from visiting its shopping site, stating that the app bypasses personalization features and violates its terms of use by capturing customer credentials and scraping account data. Meanwhile, Meta's stock surged more than 11 percent to $741 following the app's rise, and a Wall Street Journal analyst estimated Muse could add $28.5 billion to Meta by 2030. Additionally, Shopify announced a partnership with Muse to allow checkout in its online stores.
Security Vulnerabilities
Security researchers have also found a zero-day vulnerability in the Muse macOS app. Patrick Wardle, founder of the Objective-See Foundation, discovered a flaw that could potentially allow locally installed apps or terminal commands to steal user authentication tokens and hijack the AI agent. The vulnerability could let attackers intercept voice prompts processed in the cloud and inject hidden malicious commands. Meta states the app was built for privacy and security, but the software requires broad permissions such as writing files, accessing the microphone and camera, and monitoring location and calendars.
ExplainerWhy this matters
What Happened
Meta launched a personal AI agent called Muse, which quickly became the top free app on mobile app stores. The tool connects to a user's personal accounts and apps to automate daily tasks, but it was promptly blocked by Amazon over data privacy concerns and faced serious criticism from security researchers over a zero-day flaw.
Why It Matters
This situation highlights the growing clash between major tech platforms and AI companies over who controls user data and website access. As consumer AI agents try to act on behalf of users across the web, companies like Amazon are pushing back to protect their ecosystems and user credentials. At the same time, the security flaws expose the risks of giving powerful AI agents broad device permissions and cloud access.
What Happens Next
Online service providers and AI firms will likely negotiate formal partnerships, similar to Shopify's deal with Muse, to allow safe agentic commerce. Meanwhile, security researchers will continue testing AI assistants for vulnerabilities, and tech companies must patch these holes before personal AI agents see even wider adoption.
