British Startup Launches First Browser with AI-Powered Deepfake Detection

Available as a browser tool or extension, Surf can detect AI-generated deepfakes with up to 98% accuracy. How would you use this impressive technology?

A Revolutionary Tool Against Deepfakes

In a world increasingly infiltrated by artificial intelligence, the dream of a deepfake-free cyberspace is becoming a reality, thanks to London-based cybersecurity startup Surf Security.

How Does It Work?

Surf Security has developed what they claim to be the world’s first browser with a built-in feature to detect AI-generated deepfakes. This “militarized” technology employs “spatial state models” to identify AI-generated clones across various languages and accents by analyzing audio frames for inconsistencies.

As Ziv Yankowitz, CTO of Surf Security, explains, the neural network was trained using deepfakes from major vocal cloning AI platforms. He emphasizes the system’s speed and accuracy, boasting that it can “detect a fake audio track in less than two seconds.”

Ending Deepfakes

The software can discern whether audio, whether recorded or live, is produced by a human or an AI. Additionally, Surf Security plans to incorporate an AI-generated image detector into its browser soon.

AI-enabled deepfakes, which fabricate fake audios or videos, pose an increasing threat. They have been used for large-scale fraud, inciting political unrest through fake news, and damaging reputations with harmful or false content.

Surf Security has launched this deepfake detector aiming to shield businesses, media organizations, law enforcement, and military units worldwide from this escalating AI cloning threat.

Despite the challenges posed by rapidly advancing AI technology, Yankowitz admits, he assures that the full version of their deepfake detector will be available next year.