China Launches Official LLM-Powered 'National Anti-Fraud AI' App
Developed by the Shanghai Public Security Bureau under guidance from China's Ministry of Public Security, the 'National Anti-Fraud AI' mobile app and mini-programs have officially launched. The application leverages large language models (LLMs) to provide real-time Q&A, analyze fraud risks, and break down complex scam tactics for citizens. This deployment marks a high-profile, real-world application of consumer-facing generative AI tools by law enforcement to combat telecom and online fraud. It demonstrates how public safety organizations are modernizing digital fraud prevention by delivering interactive risk assessments directly to citizens' smartphones. The app offers three distinct interfaces—Standard, Elderly (featuring larger fonts and simplified navigation), and International (tailored for foreign nationals). Beyond LLM risk analysis, it incorporates anti-fraud news, video case studies, and a searchable reference dictionary with voice read-aloud support across mobile app stores, WeChat, and Alipay.
## BACKGROUND
Telecom and online fraud have become increasingly complex globally, with scammers utilizing sophisticated social engineering tactics and fake apps to deceive victims. Large language models (LLMs) enable conversational interfaces that can process unstructured user descriptions of suspicious encounters and match them against known fraud patterns.