Key Takeaways
- Meta’s new AI chatbot is sparking fresh worries about the extent of its personal data collection.
- A former employee alleged Meta identifies when users, especially teens, feel vulnerable and targets them with ads accordingly.
- The AI chatbot uses your Facebook and Instagram history, plus your chat interactions, to customize its responses and learn.
- Sources suggest this level of tracking might be more extensive than competing AI services like ChatGPT or Gemini.
- Meta’s strength lies in its ability to analyze massive amounts of user data to build detailed profiles.
- While users have some options to delete AI chat data, the overarching concern about data collection persists.
Fresh concerns about Meta’s data tracking practices have surfaced recently, fueled by its new personalized AI chat app and revealing testimony from a former employee.
Sarah Wynn-Williams, who previously worked at Meta and has written a book about her experiences, shared some unsettling details during a U.S. Senate appearance. She indicated Meta can detect when users feel “worthless or helpless,” using this as a signal for advertisers.
According to the Business and Human Rights Resource Center, Wynn-Williams stated the company informed advertisers when teens seemed depressed, allowing for precisely timed ad placements. For instance, if a teen girl deleted a selfie, this might trigger ads for beauty products, preying on potential insecurities. Similar tactics were reportedly used for weight loss ads targeting girls with body confidence issues.
The new AI chatbot from Meta also raises questions. It draws on your existing Facebook and Instagram profiles to personalize chats and meticulously tracks every interaction to refine its responses. This approach, as noted by The Washington Post, “pushes the limits on privacy in ways that go much further than rivals ChatGPT, from OpenAI, or Gemini, from Google.”
These are significant worries, though the fact that Meta possesses a wealth of information about its users is hardly new. Experts have flagged this for years. However, after the Cambridge Analytica scandal, Meta restricted data access, and the issue somewhat receded from public consciousness.
Coupled with this is the reality that many people prioritize convenience over privacy, especially if the tracking isn’t too overt. This has generally allowed Meta to avoid sustained scrutiny by simply not highlighting its tracking and Gpredictive capabilities.
Yet, numerous examples underscore the power of Meta’s user data. Back in 2015, researchers from the University of Cambridge and Stanford University published a report. It demonstrated how Facebook activity, specifically “likes,” could predict a person’s psychological profile more accurately than their friends, family, or even partners.
Meta’s true advantage here is scale. Your individual likes or page visits might seem insignificant alone. But when analyzed across millions of users, these actions can reveal strong patterns and correlations, painting a detailed picture of who you are, even if you didn’t intend to share such information.
While personal posting on Facebook has reportedly declined, potentially reducing some data points for Meta, the new AI chats offer a rich, ongoing stream of personal insights. This will undoubtedly highlight, once more, just how much Meta can understand about your preferences and leanings, which it does use for advertising.
Meta mentions in its AI documentation that “details that contain inappropriate information or are unsafe in nature” won’t be saved. Users can also delete the information Meta AI stores about them at any time, offering some level of control.
Nevertheless, this serves as a potent reminder: Meta tracks a vast amount of personal information and leverages its unmatched scale to cross-reference this data. This gives the company an enormous embedded understanding of user preferences, interests, and tendencies, all of which can be used for ad targeting, content promotion, and potentially, influence.
This is a valid concern. Yet, research consistently shows that despite available privacy settings and controls, most people don’t extensively restrict data tracking. Convenience often trumps privacy, and Meta likely hopes the same applies to its AI chatbot.
This also explains why its Advantage+ AI-powered ads are effective. As Meta’s AI tools become more sophisticated and its capacity to analyze data at scale increases, its ability to know virtually everything about you, as revealed through your Facebook, Instagram, and now AI chat interactions, will only grow. This will indeed mean a more personalized experience, but the trade-off is that Meta will also use that understanding in ways you might not fully endorse.