Artificial Intelligence (AI) robots are better at identifying images of biometric spoofing attacks compared to humans.
Indeed, a recent study by ID R&D showed that computers can spot more accurately and quickly whether a photo is of an actual person or a presentation attack. As more and more fraudsters try to imitate real customers for various processes such as accessing a bank account, the AI-based tech is able to detect liveness instantly to see if a photo, taken in real-time, is of a live person.
It was then found out that computers were more accurate than humans in tests of all five types of images (printed photos, videos, digital images, and 2D or 3D masks) with a score of 0% error rates across all 175,000 images and all types of attack. On the other hand, humans got a much lower score and degree of accuracy for every type of spoofing technique.
Moreover, AI was also 10 times quicker to identify if a photo was of a live person or a spoof. Humans needed 4.8 seconds per image to determine liveness while computers took less than 0.5 seconds per image.
This then shows the rapid rise in facial recognition for identity verification and authentication, especially in financial services where AI facial technology can detect fraud quicker.
