Inspiration Digital fraud in Thailand is growing rapidly, with scam calls, AI-generated voices, and social engineering causing billions of baht in losses every year. Many victims—especially the elderly and minors—lack awareness and technical skills, making them easy targets. AntiScam+ was inspired by the need to protect vulnerable users and transform scam evidence into actionable legal cases instead of leaving victims helpless after the incident.
What it does AntiScam+ is an AI-powered fraud prevention platform that analyzes voice conversations in real time to detect scam patterns. The system evaluates risk, compares suspicious statements with Thai legal provisions, and automatically generates structured legal documents that can be submitted to police or lawyers. This transforms a scam conversation into a complete evidence package including transcripts, timelines, and suggested legal sections.
How we built it We built AntiScam+ using a combination of AI voice analysis, behavioral risk detection, and large language models. The platform processes voice conversations, converts them into transcripts, analyzes scam patterns, and maps them to Thai legal frameworks using a retrieval-augmented architecture. The system then generates structured case documentation and presents insights through a dashboard for users, police, or legal professionals.
Challenges we ran into One major challenge was accurately detecting scam intent from real-world conversations, which often include slang, emotional manipulation, and incomplete information. Another challenge was aligning AI analysis with Thai legal terminology so that generated documents could actually assist police investigations. We also had to design the product to be simple enough for elderly users while maintaining strong privacy and compliance standards.
Accomplishments that we're proud of We successfully developed a working prototype that can analyze voice conversations, identify fraud risks, and automatically produce police-ready case documentation. The system demonstrates a full pipeline from voice input to legal output—voice → risk analysis → legal mapping → case preparation. This approach turns passive scam detection into an actionable legal intelligence system.
What we learned Through building AntiScam+, we learned that technology alone is not enough to solve fraud problems. Victims often fail to act because legal processes are complicated and evidence preparation is time-consuming. By simplifying legal documentation and automating evidence organization, AI can significantly increase the likelihood that victims report scams and pursue justice.
What's next for AntiScam+ The next step is expanding the platform beyond individual users to institutional partnerships. We plan to integrate with banks, telecom providers, and law enforcement agencies to provide early scam detection and automated case preparation tools. In the long term, AntiScam+ aims to become a national fraud intelligence infrastructure that helps prevent scams and accelerates legal prosecution across Thailand and Southeast Asia.
Log in or sign up for Devpost to join the conversation.