GovTech and Google DeepMind developing real-time detection of AI scams on devices
The model scans messaging apps on-device and warns users in real time while keeping conversations private, Google said.
- On Oct 14, Google DeepMind and the Government Technology Agency announced an on-device AI model for Android that alerts users in real-time about potential scam attempts without compromising privacy.
- Singapore has lost more than $4 billion to scams since 2019, with $410.6 million lost in the first six months of 2026 alone, driving demand for localized fraud detection solutions.
- Running entirely on-device, the technology analyzes text in WhatsApp and Telegram without sending data to cloud servers, said Google DeepMind Applied director JJ Geewax and manager Miguel Andres.
- Google also announced it will provide the Singapore Government with advanced cybersecurity tools to help defenders find and fix vulnerabilities in minutes, aligning with the Government's push for AI-powered threat defense.
- Although no release date has been disclosed, developers are exploring potential deployment in other countries, demonstrating how the Government and Google can collaborate on technology for public good.
14 Articles
14 Articles
Google Messages is Getting a Massive Gemini-Powered Scam Defense Upgrade
Google Messages is introducing opt-in Scam Detection powered by Private AI Compute and on-device Gemini Nano to catch sophisticated fraud and malicious links. Read the original post: Google Messages is Getting a Massive Gemini-Powered Scam Defense Upgrade
Google & GovTech Developing AI to Warn Users About Scam Chats Directly on Their Devices
Google DeepMind and Singapore’s Government Technology Agency (GovTech) are developing AI that could warn users about suspicious conversations on their Android devices before they transfer money or disclose sensitive information. Minister for Digital Development and Information Josephine Teo announced the collaboration at Singapore International Cyber Week on 14 October 2026. The technology is still being prepared for deployment. Google says its …
Google, Singapore's GovTech developing first-of-its-kind AI model to flag scams in real time for Android users
SINGAPORE: Google DeepMind and the Singapore Government are developing a first-of-its-kind on-device scam detection model that alerts Android users in real time to potential scam attempts.
Google DeepMind has developed with GovTech Singapore an on-device scam detection system: analyzes conversations on your device, without sending chats to cloud servers. Key points The model combines a lightweight message classifier and a custom trained Gemma model. Access to conversations requires user consent and processing remains on your device. [...]
Coverage Details
Bias Distribution
- 60% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
