Diopter

Diopter provides a platform that detects AI-enabled social engineering attacks during live calls and video meetings. It analyzes several signals at once and alerts the customer's security team while the call is in progress.

The platform runs four types of checks:

• Identity checks on each participant to flag impersonators
• Conversation analysis to identify social engineering tactics such as urgency, authority plays, isolation, and escalating requests
• Media analysis to detect synthetic (deepfake) voice and video
• Policy enforcement that flags requests violating corporate, security, and finance processes

Diopter's use cases include wire and payment fraud, help desk impersonation aimed at MFA or credential resets, and hiring and candidate fraud involving synthetic or stand-in candidates. The website also has a public deepfake detector tool.

Deployment is through a meeting bot or a desktop app. The product works with Zoom, Microsoft Teams, Google Meet, and Webex. It also works with voice and VoIP platforms such as Dialpad and RingCentral, and with carriers through an API. Customers can roll it out to devices through Intune, Jamf, Kandji, or Workspace ONE. Alerts appear in the Diopter admin console or can be sent to other tools through its API and MCP server.