DeepBrain AI launches deepfake detection solution

South Korean AI startup DeepBrain AI announced on Friday that it released the country's first service-as-a-software (SaaS)-based deepfake detection solution operating in a cloud environment.

The solution offers two paid detection services (comprehensive and voice) and supports four languages including Korean, English, Chinese, and Japanese.

The comprehensive detection service identifies manipulated images and videos, checking for features such as the application of technologies like Face Generation, Face Swap, and Lip Sync.

The voice detection service analyzes frequency and time to determine the authenticity of synthesized voices and checks for the use of AI tools like Vocoder.

DeepBrain AI highlighted the increasing misuse of deepfake technology in various crimes, emphasizing the solution's accessibility for combating these threats.

The company noted growing inquiries from domestic and international government agencies and media outlets regarding deepfake detection solutions ahead of events such as the 22nd National Assembly elections in April and the US Presidential election in November. 

Write to Kang-Ho Jang at callme@hankyung.com