AuthenticID’s New Deepfake Detection System Scans Images, Text, and Behaviors

Seattle-based AuthenticID has launched a new system designed to detect deepfakes and generative AI injection attacks. The solution is aimed at guarding against fraud during a remote identity verification process, and is fully automated.

Developed by the firm’s Product and Applied Research team, the system has three core elements. First, “visual fraud algorithms” look for synthetic media elements. Meanwhile, “text fraud algorithms” look for errors within potentially forged documents. And then there are “behavioral algorithms” that focus on the end user’s activity during the submission and capture of an ID, scanning for anomalous behavior.

In announcing the new system, AuthenticID’s SVP of Global Solutions, Stephen Thwaits, emphasized that the rapid advance of AI tools has helped fraudsters to refine their methods, requiring new and sophisticated anti-fraud solutions.

“We’ve observed fraudsters making fewer mistakes when they create fake documents,” he said. “Traditional identity verification methods can’t keep up with both the sophistication and ease at which bad actors can circumvent security measures with the use of new tools. That’s why continuous innovation is necessary to meet fraudsters at the front line.”

On that note, AuthenticID maintains an Identity Fraud Taskforce whose mandate is to continuously develop new algorithms that can detect fraud and improve the company’s identity verification decisioning engine.

Founded in 2001 by Blair Cohen, AuthenticID was initially focused on preventing document fraud, but over the years has expanded its solutions portfolio to encompass advanced AI and machine learning technologies to provide secure and accurate identity verification services. It now offers solutions across a range of sectors including telecommunications, banking, fintech, background screening, retail, e-commerce, government, healthcare, and more.

–

June 11, 2024 – by Alex Perala