AI and fingerprints: experts warn of biometric risks from selfies
EUR/MDL - 20.12 0.1556
USD/MDL - 17.32 0.3791
VMS_91 - 3.03%
VMS_364 - 9.54%
BONDS_2Y - 7.40%
GOLD - 4,580.56 1.18%
EURUSD - 1.16 0%
BRENT - 117.29 13.73%
SP500 - 733.73 0.67%
SILVER - 78.48 2.74%
GAS - 2.77 8.88%

AI has learned how to steal fingerprints directly from your selfies

Cybersecurity experts are discussing the potential threat posed by the development of artificial intelligence algorithms that could theoretically extract fingerprint elements from smartphone photos.
Natasha Kim Reading time: 1 minute
Link copied
Winston Churchill

Winston Churchill. The famous V

They estimate that images in which fingers are close to the camera (within about 1-1.5 meters) may contain enough skin surface detail to partially restore papillary lines, Techkult writes. We are primarily talking about high-resolution images, including selfies and burst photography.

Experts note that the development of computer vision methods and AI increases the potential realism of such attacks. In theory, such data could be used to attempt to bypass biometric authentication systems, including devices and payment services.

The discussion of the topic intensified after statements by cyber expert Li Chan, who claimed that modern algorithms are capable of extracting elements of the structure of papillary lines even from ordinary images. The topic spread in Chinese social networks, where the “V” gesture is often used in selfies.

That said, such research is not new. Back in 2013, security researcher Jan Krissler demonstrated the possibility of reproducing fingerprints from photographs. In subsequent years, other experiments have also shown that biometric data can be partially recovered using images and specialized equipment.

Experts emphasize that despite the potential risks, biometric authentication remains widespread due to its ease of use and high speed of access.



Реклама недоступна
Must Read*

We always appreciate your feedback!

Read also