Dubai has unveiled SARAAB, an artificial intelligence model designed to detect deepfake videos, with the Dubai Electronic Security Center (DESC) saying the system can identify manipulated content in clips as short as seven seconds.
DESC describes the model as the first open-source deepfake detection AI model developed by a government entity in the Arab region. An Emirati team at the Dubai Electronic Security Center developed it entirely.
According to DESC, SARAAB achieved up to 91% accuracy in detecting deepfake content. The model also recorded 89% accuracy when tested on previously unseen data not included in its training phase.
The center said comparative testing showed that SARAAB delivered the highest accuracy among the open-source deepfake detection models evaluated during its assessments.
The model’s ability to analyze videos as short as seven seconds is a key feature. Open-source models included in DESC’s testing generally required videos of 15 seconds or longer to perform effective analysis.
The seven-second detection capability comes as short-form videos continue to spread across social media and digital platforms. The shorter analysis requirement could make it easier to assess visual content that may otherwise be difficult to verify.
Artificial intelligence systems are increasingly capable of producing realistic fake videos, making it harder for users, organizations and platforms to distinguish authentic footage from manipulated content.
DESC said SARAAB is part of its wider efforts to develop advanced AI-driven cybersecurity solutions that keep pace with evolving digital threats and strengthen trust in the online environment.
DESC plans to make SARAAB publicly available through Hugging Face and GitHub. The release will allow AI engineers, cybersecurity professionals, and researchers to test the model, build on its capabilities, and contribute to its development.
The open-source approach is also intended to encourage collaboration across the global AI and cybersecurity community and support further development of deepfake detection technology.
Also Read: Dubai’s New AI Model Can Spot Deepfake Videos with 91% Accuracy
SARAAB was officially launched on September 16, 2026, during the opening day of GISEC Global 2026 at the Dubai Exhibition Center in Expo City Dubai.
The development comes as the UAE continues to strengthen home-grown cybersecurity capabilities. DESC said the initiative reflects its efforts to use artificial intelligence to anticipate emerging threats, protect the community and strengthen the country’s digital security.
As deepfakes become increasingly difficult to distinguish from authentic footage, tools such as SARAAB aim to strengthen the ability of users, organizations and cybersecurity professionals to verify digital content. The planned open-source release could also allow researchers and developers to contribute to the model’s continued development.