This talk will cover how to create a deep-fake social profile and that was used to exploit a target for an insurance claim and supercharge a scam. How researchers were able to impersonate the target’s voice merely by gathering a few samples of spoken audio and then synthesizing new speech. This presentation will demonstrate a step-by-step breakdown of creation.
Session recording not available per the request of the presenters.


RSA® Conference, the world’s leading information security conferences and expositions, today concluded its 31st annual event at the Moscone Center in San Francisco. The year’s event attracted over 26,000 attendees, including 600+ speakers, 400+ exhibitors, and over 400 members of the media. Throughout the week, attendees networked on the expo floor and participated in keynote presentations, track sessions, tutorials, seminars, and special events.

Several of the most pressing topics discussed during this year’s Conference included issues surrounding privacy and surveillance, the positive and negative impacts of machine learning and artificial intelligence, the nuances of risk and policy, and cybersecurity-focused innovations across crypto and blockchain.

Fake or Real, Catch Me if You Can. A Real Life Event Using Deepfake Tech

Conference: RSA Conference 2022

Abstract

Post a comment

Related work

Detecting Deep Fakes with Mice

Human or Not: Can You Really Detect the Fake Voices?

Repurposing Neural Networks to Generate Synthetic Media for Information Operations

Protecting the Protector, Hardening Machine Learning Defenses Against Adversarial Attacks