RedShift targets deepfake calls at the IP voice layer

RedShift Networks has extended the capabilities of its voice security platform to provide real-time threat mitigation and centralized security management across enterprise IP voice environments.

The company positions the platform as a zero-trust security layer against threats ranging from call spoofing and toll fraud to social engineering attacks involving AI-generated voices.

RedShift says that voice deepfake campaigns use automated botnets, spoofed caller IDs, and compromised telecommunications trunks to reach their targets.

By identifying suspicious signals, sources, and traffic behavior at the IP voice layer, the platform blocks those delivery routes before fraudulent calls reach employees or enterprise applications.

“Bad actors are exploiting voice at wire speed. We’re providing the infrastructure to stop threats before they penetrate the network,” says Mike Wagner, chief executive officer at RedShift Networks.

RedShift says employees at more than 40 percent of organizations have encountered a deepfake combined with social engineering during an audio call.

The RedShift voice security platform combines three integrated components.

Global Threat Intelligence builds on more than 15 years of proprietary behavioral data and information on more than three million known malicious telephone numbers to identify bad actors and anomalous activity.

Call Threat Manager applies that intelligence to inspect and neutralize threats before a call connects. RedShift says it has fingerprinted more than 40,000 Session Initiation Protocol attack vectors across six threat categories.

The Centralized Management System allows security teams to distribute policies, monitor system health, and enforce consistent protection across multiple locations.

RedShift says it has protected IP voice infrastructure for major telecommunications carriers for more than 15 years. The platform applies the same approach to enterprise environments by moving voice threat detection to the network layer.

“Despite the unprecedented growth in global threats, voice continues to be the only channel where identity is inferred rather than verified,” says Wagner.

Article Topics

deepfake detection  |  deepfakes  |  generative AI  |  RedShift  |  voice AI