Unverified Claims About Meta's Muse AI Assistant Raise Questions

Recent reports about Meta's new AI assistant Muse include unverified claims of a serious zero-day vulnerability and Amazon blocking the tool, but these assertions lack confirmation from primary sources or official statements.

According to the supplied text, Meta CEO Mark Zuckerberg has promoted Muse as "built from the ground up for privacy and security." The assistant, described as capable of booking appointments, filling out forms, handling customer service, making purchases, generating images, creating documents, and connecting with apps, requires extensive permissions on macOS to access user accounts and device resources.

The reported vulnerability allegedly allows any locally installed app or terminal command to access the token authenticating users to their Muse account by changing the endpoint where transcription occurs. However, the technical explanation is vague and does not logically connect to token theft, and the claim that Amazon began blocking Muse on Sunday is unsupported.

Without concrete dates, version numbers, or links to primary sources, these claims remain unverified. This article presents them as reported, not as established fact.

Source: Ars Technica