A new Android app from Singapore’s GovTech and Google DeepMind analyses incoming messages and warns users about suspicious conversations in real time. AntiScam Pango checks WhatsApp, Telegram and SMS on the phone itself, aiming to catch scams even when a sender’s number has not been reported before.
Google DeepMind Watch analysis
What happened
The app is available as a beta through Google Play. Users grant it permission to access notifications, and it runs in the background. When it flags a conversation, it explains the indicators that may suggest a scam; users decide whether to continue, report it or verify the sender another way. It does not automatically block or delete messages.
GovTech says message content is not sent to external servers unless a user chooses to report it. The app’s AI model was developed by Google DeepMind and trained on scam patterns and scenarios using data from Singapore’s ScamShield and an in-house suspicious-link classifier. The Straits Times reports that GovTech is exploring bringing the app into ScamShield, but has given no timeline.
Why it matters
Unlike tools that ask people to submit a suspicious message for checking, AntiScam Pango is designed to assess conversations as they arrive, including messages from numbers not already on a blocklist. That could give users a warning while there is still time to pause before sending money or personal information.
The practical boundaries matter too: the beta is Android-only and covers three messaging services. GovTech says it is exploring wider platform support, including iOS, but there is no launch date for that. Its privacy description is an assurance from the agency, not a reason to stop asking how on-device access and optional reporting work in practice.
Our read
This is a useful, specific application of AI: look for signs of manipulation in a conversation and explain the warning, rather than pretending a model can make the decision for you. The on-device design is central to the pitch, and the clearest test will be whether people find the alerts helpful without being overwhelmed by false alarms. If you use Android, the beta is available now; as with any new safety tool, read its permissions and decide whether they suit you.
What to watch
- Whether GovTech publishes results on detection accuracy and false alarms.
- Whether the app expands beyond Android, WhatsApp, Telegram and SMS.
- Whether AntiScam Pango is integrated into ScamShield, and on what timetable.
Discussion spark: Would you give an app access to your message notifications for real-time scam warnings, or is the privacy trade-off too much even if analysis stays on your phone?
Sources and evidence
- New GovTech-Google DeepMind app flags potential scams by analysing incoming messages – The Straits Times (4 November 2026, 04:00 UTC)
not affiliated with, endorsed by, or operated by Google or Google DeepMind