F FEICELL News全球新闻 · 本地动态 · 实时更新
← 返回新闻首页
新加坡主流

New GovTech-Google DeepMind app flags potential scams by analysing incoming messages

Read more at straitstimes.com.

The Straits TimesLee Li Ying查看原文 ↗
AntiScam Pango uses artificial intelligence to analyse all messages to flag suspicious conversations as they unfold.
AntiScam Pango uses artificial intelligence to analyse all messages to flag suspicious conversations as they unfold.

AntiScam Pango uses artificial intelligence to analyse all messages to flag suspicious conversations as they unfold.

ST PHOTO: LIM YAO HUI

Published Nov 04, 2026, 12:00 PM

Updated Nov 04, 2026, 12:19 PM

SINGAPORE – A new app has launched to help users spot potential scams by analysing and flagging suspicious conversations, including those from numbers that have not previously been reported.

Called AntiScam Pango, the app uses artificial intelligence to analyse all messages to flag suspicious conversations as they unfold.

Developed by the Government Technology Agency (GovTech) and Google DeepMind, the app was unveiled on Nov 4 by Acting Minister for Manpower Jasmin Lau at GovTech’s biennial conference STACK Conference 2026.

News of the development of this on-device scam detection model was first announced in October on the sidelines of the Singapore International Cyber Week.

“It checks messages on WhatsApp, Telegram and SMS, flags the ones that look suspicious, and warns you in real time, before you take further actions,” said Lau, who is also Senior Minister of State for Digital Development and Information and Minister-in-charge of GovTech.

“The analysis happens on your phone itself, so your messages stay private,” Lau said.

For a start, Android users can download the beta app from the Google Play Store. Once users grant the necessary permissions, AntiScam Pango runs in the background.

When AntiScam Pango flags a potentially suspicious conversation, it explains the indicators that suggest it could be a scam.

This approach aims to help users recognise social engineering tactics, in which scammers manipulate people into trusting them and taking actions that may put their money or personal information at risk.

The user can then decide whether to continue the exchange, report it or take precautions, such as independently verifying the sender’s identity. The app does not automatically block or delete messages.

Message content is not sent to external servers unless the user chooses to report it, GovTech said. Neither the agency nor its partners can access users’ personal information or private messages through the app, it added.

AntiScam Pango’s AI model from Google DeepMind was trained on scam patterns and scenarios to recognise manipulation tactics and characteristics commonly found in scams targeting people here.

Training data comes from ScamShield, developed by GovTech’s Open Government Products unit, the police and the National Crime Prevention Council.

ScamShield blocks calls from blacklisted numbers reported by users. ScamShield also uses AI to assess whether phone numbers, messages or links submitted by users are likely linked to scams. Its enhanced app has recorded 1.61 million installations since August 2024.

Training data also comes from an in-house AI-powered classifier tool that evaluates suspicious web links.

“With AntiScam Pango, we want to show people that they do not have to choose between protection from scams and keeping their personal messages private,” said GovTech’s director of trust and safety Mark Chen.

“Our collaboration with Google DeepMind is pushing the boundaries of what small AI models can do to deliver both,” he added.

AntiScam Pango adds to existing protection offered by ScamShield, which helps users block scam calls, filter suspicious SMS messages, and check or report potentially fraudulent messages, phone numbers and links.

Unlike ScamShield, AntiScam Pango proactively assesses conversations as messages arrive across supported messaging platforms, rather than rely on users to submit them for checking. It can also flag suspicious exchanges involving numbers that have not previously been identified as risky.

GovTech said it is exploring integrating AntiScam Pango into ScamShield, but no timeline has been provided.

The agency is also assessing other messaging platforms to expand Pango’s coverage beyond WhatsApp, Telegram and SMS.

For now, AntiScam Pango is available only on Android because the operating system allows apps to request access to incoming notifications from messaging services.

GovTech is exploring ways to offer similar protection on other operating systems, including Apple’s iOS, while retaining on-device analysis and the same privacy and security standards.

Li Ying is a technology correspondent covering online safety, cybersecurity and the impact of artificial intelligence.

AI/artificial intelligence

Artificial Intelligence

手机左右滑动，电脑按 ← → 键，也能切换新闻