Indonesian Developers Build AI Tool to Help Small Sellers Detect Scam Messages
A team of Indonesian developers has created SAPA (Scam Analysis and Protection Assistant), an AI agent designed to help small and medium business owners identify fraudulent messages. The tool targets two common scam types: phishing messages impersonating marketplaces and fake buyers sending forged payment screenshots to pressure sellers into shipping goods early. Rather than relying on a large language model, SAPA uses fixed pattern-matching with five weighted indicators — including OTP requests, time pressure, suspicious domains, and domain mismatches — to generate a consistent risk score in under two seconds. A message scoring 15 is flagged as high risk, while a score of zero clears a message as safe, helping prevent alert fatigue from false positives. The system also includes automatic personal data masking, replacing phone numbers, emails, OTP codes, and bank account details with tokens before any analysis is performed.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in