M83, Dblat Develop Deepfake Detection Solution 'Antigeni'

M83, a company specializing in visual effects (VFX), announced on November 4 that it had developed Antigeni, a deepfake detection solution that identifies images and videos generated or manipulated using artificial intelligence (AI), with its AI research and development subsidiary Dblat.

Antigeni determines whether content is a deepfake by analyzing visual and statistical differences between recorded footage and AI-generated content. Existing watermark detection methods have limitations when AI-generated content lacks the marks left by the generator or when those marks have been removed. Antigeni, however, analyzes the characteristics of the content itself, enabling it to detect deepfakes generated by a variety of AI models.

Development has also been completed on browser extensions for Google Chrome and Safari on iPhone. Users can view analysis results for images and videos while browsing the web.

The two companies plan to improve detection accuracy and expand coverage to address new generative AI models. The browser extension will be offered to individual users as a free or paid service, depending on its performance. They also plan to supply AI detection engines to media outlets, content platforms, and public institutions, and provide customized authenticity verification solutions for clients.

An M83 representative said, "We will pursue the commercialization of Antigeni and secure new business opportunities by establishing various revenue models, including supplying application programming interfaces (APIs) and licensing enterprise solutions."

This content was produced with the assistance of AI translation services.