Meta’s Muse is creepy, but maybe not for the reasons you think
Meta's Muse AI assistant appeared creepy due to message access issues, but it was a misunderstanding.Summary
In an article by Terrence O'Brien for The Verge, Meta's Muse AI assistant is described as effective but creepy, particularly after an incident where it seemed to access user messages without explicit permission. Contributing editor Jason Aten shared screenshots on Threads showing Muse referencing a conversation from his Messages app, despite claiming he hadn't granted access. Muse initially explained it saw notification previews, but when pressed, it couldn't clarify how it received them. Meta's David Singleton later clarified that Muse syncs data only after user-enabled access and apologized for Muse's incorrect response, attributing it to the AI's confusion about its own features. This incident highlights concerns about AI transparency and the tendency for chatbots to provide inaccurate explanations.
(Source：The Verge)