Source: Google BlogMay 28, 2026

Gemini Visual AI Triggers Launch on Nest Cameras

View original source →

Google launched Gemini visual AI triggers for Nest cameras on May 28, replacing rule-based motion detection with semantic scene understanding.

How it works:

• Users define natural-language conditions like 'alert when a delivery arrives on the porch' • Gemini evaluates conditions against live camera feeds in real time • On-device inference for privacy-preserving processing • Cloud-based Gemini Pro for complex multi-condition triggers exceeding device capacity

Integration and privacy:

• Covers indoor and outdoor Nest camera lines via firmware update • Integrates with Google Home automation routines • Triggers actions across lighting, locks, thermostats, and Matter-compatible devices • Visual trigger history stored locally, not used for model training without explicit consent

Why It Matters: This is the first mass-market deployment of a frontier-class vision model as the primary interface between physical environments and digital automation — AI as sensory layer, not just assistant.