AI More Accurate Than People at Identifying Biometric Spoofs, New Study Shows

Rhea-AI Summary

The latest research by ID R&D reveals that computers outperform humans in detecting biometric spoofing attacks, such as photos and videos. The AI systems achieved a 0% error rate across 175,000 images, while humans misidentified 30% of printed images. Machines were nearly 10 times faster, taking less than 0.5 seconds per image compared to 4.8 seconds for humans. The study emphasizes the effectiveness of AI-based facial liveness detection in identity verification, significantly reducing false positives for genuine users to just 1%.

Positive

• Machines achieved a 0% error rate on 175,000 images in detecting spoofing.
• AI systems were nearly 10 times faster, processing images in under 0.5 seconds.
• Only 1% of genuine faces were misclassified as spoofs by AI, compared to 18% for humans.

Negative

• None.

Machines also 10X faster at detecting fake faces, important to curb online ID fraud

NEW YORK--(BUSINESS WIRE)-- Humans have far greater difficulty identifying images of biometric spoofing attacks compared to computers performing the same task, according to research released today by ID R&D, a provider of AI-based voice and face biometrics and liveness detection technologies.

The new research report, Human or Machine: AI Proves Best at Spotting Biometric Attacks, finds that computers are more adept than people at accurately and quickly determining whether a photo is of an actual, live person versus a presentation attack. Fraudsters attempt to imitate real customers during processes such as creating a new bank account or logging into an existing account. Liveness detection instantly validates whether a photo, taken in real time, is of a live person.

The study tested humans and machines by presenting them with the most common spoofing techniques: printed photos, videos, digital images, and 2D or 3D masks.

Machines outperformed humans for all types of face biometric spoofing

Computers were more accurate than humans in tests of all five types of images, scoring 0% error rates across all 175,000 images and all types of attack. Humans had a far lower degree of accuracy for every type of spoofing technique, including misidentifying 30% of photo prints, one of the easiest attack types for fraudsters to execute. Even when a group of 17 people voted on the images, resulting in a more accurate outcome than an individual person, their majority decisions were never better than the computer’s performance of the same task.

Computers were also almost 10 times quicker to recognize a photo of a live person or a spoof. On average, it took humans 4.8 seconds per image to determine liveness, whereas computers running on a single CPU took less than 0.5 seconds per image to determine liveness. These latest technology advances support the rapid rise in facial recognition for identity verification and authentication.

This performance is strong evidence for organizations in financial services and other industries staking trust in automation. The ability to use AI facial liveness technology to detect fraud saves time and enables human resources to focus on more complex fraud.

Technology ensures the most frictionless customer experience

Despite the strong performance of computers at spotting spoofs, fraud detection must not compromise the experience of genuine customers. Many facial liveness systems on the market are good at keeping fraudsters out, but in the process, a significant number of genuine people are also caught in the net.

However, in this study, the AI system erroneously classified just 1% of genuine faces as spoofs. Humans, on the other hand, misclassified 18% of genuine faces as spoofs, confirming that passive facial liveness detection is also better than humans at keeping genuine customers out of the fraud net.

“The results are undeniable,” said Alexey Khitrov, CEO at ID R&D. “Biometric technology used for identity verification has evolved in recent years to increase speed and accuracy, now significantly outperforming the human eye. Organizations can achieve tremendous efficiencies by using identity verification systems that include a biometric component. However, there is still work to be done and we are excited to see biometrics helping to build consumer trust.”

Notes to Editors

The new research report, titled Human or Machine: AI Proves Best at Spotting Biometric Attacks is available now.

About ID R&D
ID R&D, a Mitek company, is an award-winning provider of AI-based voice and face biometrics and liveness detection. With one of the strongest R&D teams in the industry, ID R&D consistently delivers innovative, best-in-class biometric capabilities that raise the bar in terms of usability and performance. Our proven products have achieved superior results in industry-leading challenges, third-party testing, and real-world deployments in more than 50 countries. ID R&D’s solutions are available for easy integration with mobile, web, messaging, and telephone channels, as well as in smart speakers, set-top boxes, and other IoT devices. ID R&D is based in New York, NY.

About Mitek
Mitek (NASDAQ: MITK) is a global leader in mobile capture and digital identity verification built on the latest advancements in computer vision and artificial intelligence. Mitek’s identity verification solutions enable organizations to verify an individual’s identity during digital transactions to reduce risk and meet regulatory requirements, while increasing revenue from digital channels. More than 7,500 organizations use Mitek to enable trust and convenience for mobile check deposit, new account opening and more. Mitek is based in San Diego, Calif., with offices across the U.S. and Europe. Learn more at www.miteksystems.com.

View source version on businesswire.com: https://www.businesswire.com/news/home/20220209005062/en/

Angela M. Romei

Corporate Communications Director

pr@miteksystems.com



Sarah Hunt

LEWIS for Mitek

MitekUK@teamlewis.com

Source: ID R&D

FAQ

What does the new ID R&D research reveal about biometric spoofing detection?

The research shows that computers significantly outperform humans in detecting biometric spoofing, achieving a 0% error rate.

How much faster are machines at detecting liveness compared to humans?

Machines are nearly 10 times faster, processing images in less than 0.5 seconds, while humans take an average of 4.8 seconds.

What is the false positive rate for AI in detecting genuine faces?

The AI system misclassified only 1% of genuine faces as spoofs, demonstrating high accuracy.

What are the implications of this research for identity verification systems?

The findings support the use of AI in identity verification, enhancing efficiency and reducing fraud while improving customer experience.

How does this research impact Mitek's position in the market?

Mitek, as a leader in digital identity verification, can leverage these findings to enhance its biometric technology offerings.