AI's New 'Muse': Friend or Digital Spy in Your Personal Life?
AI's New 'Muse': Friend or Digital Spy in Your Personal Life?
Meta's Muse promises to organize your social world, but its deep dive into your relationships raises serious privacy questions. Is the convenience worth the cost?
Muse: Your New Best Friend or a Digital Dossier?
Meta's new AI assistant, Muse, has exploded onto the scene, charming millions into downloading it and connecting it to their most intimate data: bank accounts, messages, even health records. The promise is alluring: an AI that can anticipate your needs and streamline your life. But as the curtain is pulled back on Muse's inner workings, a different, more unsettling picture emerges, one that forces us to question the true cost of convenience.
Recent discoveries by researchers, including independent AI safety and security expert Karan Joshi, reveal Muse's instructions for creating deeply personal profiles. Imagine “a page for every person in the user's life”, yes, literally. This hourly process diligently compiles data on your family, partners, friends, colleagues, and even people you just follow.
It's designed to build a 'memory' of your relationships, using structured text files to suggest everything from relationship improvements to coffee spots for a friend.
While AI chatbots generally track social information, Muse's specific architecture, combined with Meta's vast access to social network data, raises red flags. Joshi observes, “What it seemed like to me—from all these prompts, system skills data, and things that they’re feeding into Muse—is that they want to understand your relationships that you have with real people.” He adds, chillingly, “They’re trying to know you like a friend, which is honestly pretty creepy.”
These “pages” are no mere contact lists. Muse's documentation outlines sections like 'Facts,' 'History,' 'The relationship,' 'In common,' 'Open threads,' and 'Strengthening.' It aims to record details such as where people live, what they do, recurring conversation threads, and significant dates like birthdays and anniversaries.
The 'History' section could document past events, like “the trip in March, the argument that got resolved, the milestone last week.” The 'Strengthening' section even suggests ways to improve your connections, offering prompts like “A reason to call, a date worth remembering, something they said to circle back on.”
The intent seems clear: to create an incredibly detailed, dynamic social graph of your life. But at what point does a helpful assistant become an omnipresent monitor?
As Carissa Véliz, an associate professor at Oxford's Institute for Ethics in AI, aptly puts it,
We are giving AI systems much more information about us than we are getting information from them.
This isn't just about what we explicitly share, but what Muse can infer, correctly or incorrectly, and piece together from various data sources.
Is This the Future We Want?
The integration of AI into our daily lives is inevitable, but the terms of engagement are still being written. Muse offers a glimpse into a future where an AI could become an intimate confidante, meticulously documenting the fabric of our social existence. The question isn't whether this technology can exist, but whether it should exist in this form.
Are we willing to sacrifice such profound levels of personal data and relational intimacy for the promise of streamlined social interaction? It's a trade-off we, as users, must actively consider before handing over the keys to our most cherished connections.
