Onfido releases Global Fraud Index and Socure commits to eliminate digital identity fraud ahead of Money 20/20

Categories Biometrics News
Onfido releases Global Fraud Index and Socure commits to eliminate digital identity fraud ahead of Money 20/20

Roughly 1 percent of $5 trillion a year in global fraud is recovered by authorities, according to Onfido’s inaugural Global Fraud Index. The company has also announced it will demonstrate new machine learning capabilities at the 2019 Money 20/20 USA.

Roughly one in three U.S. adults have been affected by identity theft, according to a blog post announcing the report. The report also found that fraud incidents mostly occur during weekdays, as criminals take a professional approach to attacks.

The report also analyzes the rate of fraud for certain types of ID documents presented. The most commonly fraudulent document was Russian driving licenses, 18 percent of which were suspected to be fraudulent. Vietnamese national ID cards were suspected to be fraudulent in 12 percent of cases, the next most, followed by Ukrainian driving licenses (7.6 percent). The least fraud was suspected among Russian passports (4 percent), while U.S. passports were suspected frauds 6.1 percent of the time, roughly average.

At Money 20/20, Onfido will show new capabilities for monitoring the quality of identity document data capture, enhancing images to minimize the number of cases requiring human intervention or applicant re-submission, an algorithmic model to recheck the system’s own work to ensure sufficient confidence, and decision logs with a heat map over the captured document image for improved auditability.

“Biometrics have played an important role in digital commerce, online banking, and other parts of the FinTech market, with AI and machine learning being a key driver in its growth,” said Husayn Kassai, CEO and co-founder of Onfido. “As we’ve highlighted in our Fraud Index, identity issues are critical and core to the financial services industry; as offerings continue to evolve, the ability to reduce friction and enable greater access while increasing trust is our key focus.”

Kassai and Onfido Chief Product Officer Kevin Trilli will both participate in panel discussions at Money 20/20 as well.

Socure says it plans to eliminate third-party identity fraud, and will introduce three new key product enhancements to its Socure ID+ platform at Money 20/20 to support this goal.

“Powered by artificial intelligence and machine learning, our platform is taking identity verification capabilities to the next level. Already, we have customers capturing 100 percent of the identity fraud found in the riskiest 3 percent of the population,” said Tom Thimot, CEO of Socure. “These enhanced versions of our highly trusted products will get us one step closer to completing our identity data set and officially eliminating digital identity fraud in the U.S.”

The latest version of the ID+ Identity Verification Platform, version 3.0, includes sigma fraud scoring with machine learning and artificial intelligence, which the company says reduces manual review time by up to 90 percent, as well as document verification, which combines with biometric selfie and liveness detection, and KYC/CIP checks with proprietary ML techniques for auto-approval rates higher than 90 percent.

“When we started Socure, we were told that we could never reach higher than 90 percent AUC, our aggregate measure of anti-fraud performance across all possible classification thresholds,” said Pablo Abreu, SVP of Data Science at Socure. “We took that as a challenge. Now we are reaching the inevitable accuracy of moving from a probabilistic to a deterministic system. Today, we are at 97.2 percent AUC and are steadily improving 1.5-2.0 percent AUC every six months. Once our accuracy reaches 99.9 percent AUC, identity fraud will essentially be eliminated as an issue for our customers.”

Both companies were recently honored for their innovation in digital identity verification.

Money 20/20 is being held October 27 to 30 in Las Vegas.

Related Posts

Article Topics

 |   |   |   |   |   |   |   | 

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Most Read This Week

Featured Company

Biometrics Research

Biometrics White Papers

Biometrics Events

Explaining Biometrics