// THREAT DETECTION AND DATA PRIVACY TERM
Deepfake
A deepfake is synthetic media, like a video or audio recording, created using artificial intelligence to realistically replace someone's likeness or voice with another's. They are often used to create convincing but fake content for scams, misinformation, or harassment.
TECHNICAL DEFINITION
A deepfake is a form of synthetic media generated by deep learning algorithms, such as generative adversarial networks (GANs) or variational autoencoders (VAEs), to manipulate or fabricate audio-visual content. In cybersecurity, this AI-driven threat is used for social engineering, disinformation campaigns, identity theft, and sophisticated phishing attacks (vishing) by creating realistic impersonations of individuals.
BACKGROUND
Deepfakes are images, videos, or audio that have been edited or generated using artificial intelligence, AI-based tools or audio-video editing software. They may depict real or fictional people and are considered a form of synthetic media, that is media that is usually created by artificial intelligence systems by combining various media elements into a new media artifact.
READ MORE ON WIKIPEDIASYNONYMS & ALIASES
- synthetic media
- AI-generated media
- face swap
- voice cloning
- digital forgery
- generative adversarial network output
USAGE NOTE
In a security context, deepfakes are a growing concern for voice phishing (vishing) attacks, where an attacker might clone an executive's voice to authorize fraudulent wire transfers.
DEVELOPERS
Organizations developing technology related to Deepfake.
Intel has developed FakeCatcher, a real-time deepfake detection technology that analyzes blood flow in video pixels to determine authenticity. It looks for subtle physiological cues that are difficult for AI to replicate.
The Defense Advanced Research Projects Agency (DARPA) runs programs like Semantic Forensics (SemaFor) and Media Forensics (MediFor) to develop advanced technologies for automatically detecting, attributing, and characterizing manipulated media, including deepfakes.
Microsoft has developed the Video Authenticator tool, which can analyze a still photo or video to provide a confidence score on whether the media is artificially manipulated. They are also a co-founder of the Coalition for Content Provenance and Authenticity (C2PA).
A company offering a proactive detection platform for enterprises and governments to identify deepfakes across various media types, including audio, video, and images. Their solution helps prevent fraud, disinformation, and security breaches.
Truepic develops technology to verify the authenticity of digital photos and videos from the point of capture. Their controlled capture technology creates secure, signed media that can be used to prove its origin and lack of manipulation, directly countering deepfakes.
Sensity provides a threat intelligence platform that detects and monitors synthetic media, including deepfakes. They help organizations identify and respond to risks posed by visual and audio fakes, such as brand impersonation, fraud, and extortion.
iProov specializes in online biometric face authentication. Their technology includes 'liveness assurance,' which is designed to defend against sophisticated attacks using deepfakes and other digital spoofs to ensure a user is a real person, present right now.
As a leader in digital media, Adobe co-founded the Content Authenticity Initiative (CAI), which focuses on creating a secure, end-to-end system for providing provenance and attribution for digital content, serving as a tool against deepfakes.