mAIScam

Overview

Southeast Asia faces a $64 billion cybercrime crisis, with scammers exploiting the region’s linguistic diversity to launch sophisticated phishing attacks, fake websites, and social media fraud in 13+ local languages.

Current security tools are reactive and culturally blind, leaving vulnerable populations—especially the elderly—exposed to devastating financial losses from threats specifically designed to exploit the region’s unique linguistic and cultural landscape.

Solution:

mAIscam is an AI-powered web browser extension built on SEA-LION v4 that delivers real-time, culturally intelligent scam protection across emails, websites, and social media in 13 Southeast Asian languages. It acts as a 24/7 digital guardian that understands regional scam patterns and linguistic nuances.

How It Works:

Through our web browser extension, potential threats are detected, captured and sent to our backend (powered by AWS), where SEA-LION v4’s multimodal AI analyses email content, social media posts (text and images), and website metadata. It cross-references URLs against the PhishTank database, validates contact information, and checks SSL certificates, delivering clear risk assessments (High/Medium/Low) with culturally-appropriate explanations in the user’s native language within seconds.

mAIscam Detection Result

Key Features:

• Email & Website Protection: Scans Gmail for phishing attempts and validates websites for SSL certificates and brand impersonation
• Social Media Analysis: Evaluates Twitter and Facebook posts through text, images, and engagement patterns
• Active Protection: Automatically blocks flagged websites, blurs suspicious content, and enables one-click scam reporting
• 13 Languages: Delivers threat assessments in Southeast Asian languages with culturally-tailored explanations

mAIscam Active Protection – blurring the detected scam social media post

What Makes It Innovative:

mAIscam leverages SEA-LION v4‘s native understanding of Southeast Asian languages and cultural contexts through prompt engineering. As a browser extension, users receive seamless protection without leaving their browsing environment. An admin dashboard enables real-time threat monitoring for organizations and governments to track emerging regional scam patterns.

Overview of mAIscam features

Conclusion:

During the Pan-SEA AI Developer Challenge 2025, we developed a fully working prototype and won 2nd Place in the Finance Track. User feedback has validated mAIscam’s technical strength and accessibility, while highlighting key areas for improvement: enhancing detection accuracy, reducing inference delays, implementing auto-detection for incoming emails, and developing a mobile app version.

We are actively refining the solution and in discussions with AI Singapore for potential research collaboration to enhance SEA-LION’s scam detection capabilities. We’re also seeking partnerships with local financial institutions, telecommunications companies, and governments to validate real-world use cases, demonstrate measurable impact, and build the trust necessary to scale across the region.

Together with the SEA-LION model, we are building a safer, more secure online environment for everyone in Southeast Asia—protecting vulnerable communities from cyber threats while empowering digital trust and financial inclusion across diverse languages and cultures.

About the team:

mAIscam is built by a team from the Singapore Institute of Technology (SIT), driven by personal experience with scam victims in their own families.

Team Members:

Shaun Liew – AI Research Engineer at SIT x NVIDIA AI Centre specializing in Visual Language Models and AI-Generated Content. In this project, he serves as an AI and Cloud Engineer, bringing expertise in Python, PyTorch, and Computer Vision.

Edmund Lin – Final-year SIT student and Blockchain & AI Enthusiast with expertise in LLMs, blockchain, and full-stack engineering. He serves as a full-stack engineer for the project.

Our Motivation:

We’ve witnessed firsthand the devastating financial and emotional impact when our own family members fell victim to scams. Knowing we’re not alone in this experience, we built mAIscam to ensure others don’t suffer the same painful lessons—creating a solution that protects vulnerable communities across Southeast Asia from sophisticated cyber threats.

mAIscam Team Photo