Meta, the parent company of Facebook and Instagram, is bringing back facial recognition technology in a bid to combat fraudulent “celeb-bait” ads and improve account recovery options. After discontinuing the use of facial recognition in 2021 due to privacy concerns, Meta is now reintroducing it in a more targeted form. This initiative is focused on detecting scam advertisements that misuse the likeness of public figures, often through the use of deepfake or AI-generated imagery.
The Problem with “Celeb-Bait” Ads
Scam ads that falsely feature celebrities endorsing products or investment schemes have been a persistent issue on social media platforms. These ads lure users into clicking, only to lead them to fraudulent websites. Celebrities such as Tom Hanks and Oprah Winfrey have repeatedly issued statements disassociating themselves from these misleading ads, yet the scams persist.
Meta solution involves using facial recognition to compare faces in suspected scam ads with the profile pictures of public figures on Facebook and Instagram. If a match is detected and confirmed to be fraudulent, the ad will be blocked from appearing on the platforms. The tool is currently being tested with a small group of public figures and celebrities, and Meta plans to expand its use in the coming months.
Enhancing Account Recovery
In addition to fighting scam ads, Meta is testing a “video selfie” feature to assist users in recovering compromised accounts. The feature will allow users to verify their identity by uploading a video of themselves, which will be compared against their profile pictures. This method is expected to help users regain access to their accounts more securely and prevent unauthorized access by scammers.
Meta has emphasized that the facial data used in these processes will be immediately deleted after comparison, alleviating some privacy concerns. However, the reintroduction of facial recognition is not without controversy, especially given Meta’s history with biometric data. Notably, the new features will not be available initially in regions with strict privacy laws, such as Illinois, Texas, the UK, and the European Union. Talks are ongoing with regulators in these regions to potentially roll out the technology globally by 2025.
Striking a Balance Between Security and Privacy
The reintroduction of facial recognition highlights Meta ongoing struggle to balance security measures with user privacy. While this move may significantly reduce the spread of scam ads and improve user security, it is likely to reignite debates over the ethical use of biometric data on social media platforms. Despite these concerns, Meta seems committed to using cutting-edge technology to protect its users from online fraud while being cautious about data privacy.
Monika Bickert, VP of content policy at Meta, has stated that “protecting our users from online scams is a top priority,” adding that the company is investing heavily in technologies like facial recognition to help identify and block fraudulent content before it reaches users.
As this technology continues to roll out, Meta will likely face ongoing scrutiny over its impact on user privacy. However, for many users particularly those who have been victimized by scam ads or account hacks the promise of more robust security measures may be worth the trade-off.
However, it seems that this feature cannot help everyone, especially business accounts that do not have a person’s profile photo. Previously, privacy was the reason Meta turned off this feature three years ago, but they have now promised to immediately delete the facial data used and ensure that the data is not used for other purposes.
