Ping An Digital Bank, a Hong Kong-based virtual lender under Ping An, said on October 15 that it had completed an anti-fraud project in the Hong Kong Monetary Authority’s second generative AI sandbox cohort to test tools for detecting deepfake fraud.

The Anti-Fraud Strategy Platform, developed with OneConnect Financial Technology, uses facial recognition during identity verification to spot manipulated or synthetic selfies in real time, according to Ping An Digital Bank. The bank said sandbox tests showed detection accuracy of more than 96%.

Ping An Digital Bank also said the project explored an “AI vs. AI” approach, in which AI systems identify AI-generated fraud. In a separate 2024 Hong Kong case, a finance worker at a multinational company was duped into transferring US$25 million in a video-call scam.

🔗 Source: Ping An Digital Bank

Recent Ping An Digital Bank developments