FaceApp Privacy Fears Miss the Mark on Facial Recognition
Despite widespread panic about the viral aging app FaceApp, experts note the company is highly unlikely to use uploaded photos for facial recognition due to a lack of identifying information and poor single-photo accuracy.
The viral photo-editing app FaceApp faces intense public scrutiny and calls for an FBI investigation over its data practices. The app, which uses artificial intelligence to alter photos by aging users or changing their appearances, gathers millions of face photos under a broad terms of service agreement. Because the app is owned by a Russia-based company, users and politicians express serious concern that this massive database could be exploited for surveillance or facial recognition tracking.
However, technology experts clarify that using these photos for facial recognition is highly impractical and unlikely. Most users do not provide their real names or identifying details to the app, which makes it incredibly difficult to link a face to a specific identity. Additionally, while AI can theoretically learn to recognize a person from a single edited photo, the accuracy remains very poor, making social media profiles a much easier and more reliable source for tracking individuals.
Instead of building recognition tools, the company most likely uses the uploaded images to improve its own core image-editing algorithms. Feeding a continuous stream of new face photos into its neural network allows FaceApp to fine-tune its ability to modify and alter facial features. This type of dataset holds much more value for advancing the app's generative capabilities than for creating an inaccurate and anonymous surveillance system.