SoylentNews
SoylentNews is people
https://soylentnews.org/

Title    Phones' Facial Recog Tech 'Fooled' by Low-Res 2D Photo
Date    Wednesday May 24, @04:19PM
Author    janrinok
Topic   
from the dept.
https://soylentnews.org/article.pl?sid=23/05/23/1831213

janrinok writes:

Someone who looks a lot like you could also unlock it, says Which?

Samsung, Oppo and Nokia are among a range of Android phone makers with facial recognition scanning tech that can be "easily duped" by a printed 2D photo, according to tests undertaken by campaign group Which?

Resident techies that put a range of phones and brands through their paces (see box below) said the findings were of concern as biometric tech is often billed as one of the most secure ways to unlock a handset.

Of the 48 phones Which? sent to labs for testing, 19 could be spoofed with photos and "worryingly" these were "not even particularly high resolution and were printed on a standard office printer on normal, rather than photo, paper."

The vast majority of the phones that failed the simple biometric test were, unsurprisingly, low to mid-range in price, though Which? claimed there were exceptions, including the Xiaomi 13 and the Motorola Razr.

Of the phones that Which? reckons could be fooled, seven were made by Xiaomi, four came from Motorola, while two came from each of Nokia, Oppo and Samsung. One model made by Honor and another by Vivo was also found to be exploitable.

Under Android's requirements, phone makers must ensure devices and software are "Android compatible," which includes how often device security can be spoofed. Class 3 systems must not be duped more than 7 percent of the time, and Class 1 system are least secure, with a spot rate of 20 percent of the time to more.

Which? voiced worries that scammers could exploit the weakness to – for example – access Google Wallet to make payments to a limited value (£45 in the UK, about $56) without needing to unlock their phone. For larger transactions, Google asks users to use a Class 3 biometric lock, Which? said.


Original Submission

Links

  1. "janrinok" - https://soylentnews.org/~janrinok/
  2. "Someone who looks a lot like you could also unlock it, says Which?" - https://www.theregister.com/2023/05/19/2d_photograph_facial_recog/
  3. "Original Submission" - https://soylentnews.org/submit.pl?op=viewsub&subid=59707

© Copyright 2023 - SoylentNews, All Rights Reserved

printed from SoylentNews, Phones' Facial Recog Tech 'Fooled' by Low-Res 2D Photo on 2023-06-03 14:54:26