Ensign InfoSecurity launches real-time deepfake detection tool

Asia's largest cybersecurity provider, Ensign InfoSecurity, has introduced Aletheia, a pioneering real-time deepfake detection solution developed in Asia-Pacific (APAC). This innovative tool aims to identify AI-driven manipulations in audio and video content within a second, achieving a detection confidence rate of 90%. Developed by Ensign Labs, Aletheia will be commercially available as both endpoint software and a Chrome browser plug-in starting next month. A mobile version is also under development.

The growing global threat posed by AI-generated deepfakes has motivated this timely solution, addressing a significant vulnerability affecting enterprises and governments. Deepfakes have increasingly been used by cybercriminals for activities like phishing and fraud. Highlighting the urgency, Singapore recently introduced a Bill to restrict the use of deepfakes and other AI technologies that could distort electoral outcomes. Aletheia's real-time detection capabilities can help prevent such manipulations, thus maintaining content integrity and protecting public trust.

Deepfake detection often relies on identifying visual irregularities such as awkward facial positions, inconsistent body movements, flickering, and mismatched lip movements. However, advancements in generative AI and large language models have made it easier to produce highly realistic fake audio and video content. This poses significant risks to various sectors, including finance, media, and public services. Aletheia enables users to discern between real and fake content in real time, providing an invaluable tool against such threats.

Ensign Labs' development process for Aletheia involved using open-source tools to generate both deepfake and authentic audio and video content. These datasets were subsequently used to train their suite of AI detection models. The Aletheia solution uses AI algorithms to analyse video pixels and audio frequencies in the background, providing authenticity notifications within seconds and with up to 90% accuracy.

Available as an endpoint software for Windows operating systems, Aletheia performs AI analysis and detection locally on the user's device, ensuring both privacy and minimising latency issues. For non-Windows systems, Aletheia will also be available as a Chrome browser plug-in. The solution can also integrate with an organisation's Security Information and Event Management (SIEM) systems, giving enterprises a more comprehensive view of their digital security posture.

"The democratisation of deepfake technology has raised serious societal risks, from commercial frauds and the spread of falsehoods to the manipulation of public opinion to discredit politicians," said Mr Tan Ah Tuan, Head of Ensign Labs at Ensign InfoSecurity. "As the technology landscape evolves, the arms race between creating and detecting deepfakes will only intensify. This also means that Aletheia will be continuously updated to combat the latest deepfakes. As cyber defenders, Ensign InfoSecurity strives to be at the forefront of innovation, developing the right solutions for enterprises and individuals to navigate the complex digital landscape," he added.

Ensign InfoSecurity will showcase Aletheia and other AI-powered cybersecurity solutions at the upcoming GovWare 2024 event, under the theme "AI-Powered Innovations: Safeguarding Your Reputation". The event will take place from 15 to 17 October at the Sands Expo and Convention Centre.