DeepMind and Singapore's GovTech launch on-device anti-scam model
What happened
On October 14, Google DeepMind and Singapore's Government Technology Agency (GovTech) released an on-device scam detection model. Built on a fine-tuned lightweight Gemma, it runs locally on ordinary smartphones to spot scam threats in messages and conversations. Detection uses a two-stage pipeline: a lightweight classifier screens messages first, then the fine-tuned Gemma analyzes conversation context to confirm the threat and trigger a real-time warning. All data is processed on the device to protect user privacy.
Written by AI from the coverage · updated 2 hours ago
Coverage
Follow the reports to see the story from different sides.
- Google DeepMindGoogle DeepMind and Singapore Government announce new AI tool to protect citizens against scams
Google DeepMind 与新加坡政府科技局（GovTech）合作推出端侧诈骗检测模型，基于轻量级 Gemma 微调，可在普通智能手机上本地运行。该模型采用两阶段流水线：轻量分类器先筛查消息，再由微调 Gemma 分析对话上下文确认威胁并触发实时警告，全程在设备端处理数据以保护隐私。
Heat over time
Not enough continuous observations to draw a trend yet.