Dubai has introduced a new artificial intelligence model designed to detect deepfake videos and identify digitally manipulated content with an accuracy rate of 91 percent.
The Dubai Electronic Security Centre (DESC) unveiled the AI-powered system, named SARAAB, at GISEC Global 2026 in Dubai, highlighting the growing use of artificial intelligence to address emerging digital security threats.
According to DESC, SARAAB has been developed to detect manipulated video content and help counter impersonation threats affecting both public and private sector organizations.
The centre described SARAAB as the first deepfake detection model in the Arab world to be developed entirely by a government agency. The system is intended to strengthen efforts to identify fraudulent or altered digital media.
An Emirati team developed the artificial intelligence model in just six months, according to the Dubai Electronic Security Centre. The development reflects efforts to build locally developed technologies for addressing cybersecurity and digital manipulation challenges.
SARAAB analyzes complete video clips rather than examining only selected frames. The system is designed to identify signs of digital manipulation throughout the footage and provide visual indicators of potentially altered sections.
One of its key features is the use of heat maps that display moving hotspots. These visual indicators can help show which frames or parts of a video have been identified by the system as potentially manipulated.
The technology could have applications across organizations that need to verify the authenticity of video content. Deepfake technology has become increasingly sophisticated, allowing digitally altered videos to imitate real people and create misleading visual material.
The ability to detect manipulated footage has therefore become an important part of modern digital security. Government agencies, businesses and other organizations can face risks when altered videos are used for impersonation, misinformation or fraudulent activity.
With SARAAB, DESC is positioning artificial intelligence as a tool for detecting such threats and supporting the verification of digital content.
The announcement at GISEC Global 2026 also highlights Dubai’s broader focus on developing and deploying advanced cybersecurity technologies. The 91 percent reported accuracy rate gives the system a measurable benchmark for its deepfake detection capabilities.
As AI-generated and digitally manipulated videos continue to develop, technologies capable of identifying altered content are expected to become increasingly important for digital security and content verification.
