Google Home adds Gemini-powered camera automation for smart home routines

At a glance:

• New Gemini-powered automation triggers smart home routines based on camera visuals.
• Supported on Nest and select third-party cameras; requires Google Home Premium subscription.
• Enhanced voice commands allow multiple actions in a single request.

What's new in Gemini for Home

Google Home is rolling out a significant update to its Gemini-powered automation suite, enabling users to create smart home routines triggered by visual insights from security cameras. This feature, announced alongside improvements to voice command consistency and system stability, represents a step forward in integrating AI-driven perception into everyday home management. The update follows the early access launch of Gemini for Home in October, which laid the groundwork for more intuitive smart home interactions.

The core innovation lies in allowing cameras to "understand" what they see, translating visual data into actionable triggers. For instance, users can configure routines to activate when specific objects, vehicles, or even people (via the Friendly Faces feature) are detected. However, Google cautions against relying on this for time-sensitive scenarios, emphasizing that processing delays make it unsuitable for critical safety or security alerts.

How the camera automation works

To set up the feature, users describe the desired event in natural language, such as "raccoons near the trash bins" or "red BMW enters driveway," and select which cameras should monitor for it. The system processes visual input to identify matches, then executes predefined routines like turning on lights or adjusting thermostats. Google recommends focusing on clearly visible objects to ensure reliable detection, as the camera requires a brief moment to analyze scenes.

Examples include triggering security lighting to deter wildlife, receiving notifications when mail is delivered, or activating indoor routines like opening smart blinds when a specific car arrives. These capabilities depend on enabling AI descriptions through the "Gemini for Home camera features" setting, which must be activated for compatible devices.

Voice command enhancements

Gemini for Home now supports multi-action voice commands, allowing users to bundle tasks like lowering blinds, dimming lights, setting timers, and playing podcasts in a single request. The voice assistant also responds faster and provides more predictable outcomes when interpreting natural language inputs, such as adjusting room lighting to be "a little warmer."

These improvements aim to streamline user interaction with the platform, reducing the need for repetitive commands. The update reflects Google's broader push to make smart home ecosystems more conversational and less reliant on rigid, pre-programmed phrases.

Availability and requirements

The camera automation feature is currently limited to US users enrolled in the Google Home Public Preview program and requires a subscription to the Google Home Premium Advanced plan ($20/month or $200/year). Compatible devices include Nest cameras and select third-party models with "Gemini Built-In" capabilities, though Google has not specified which third-party brands qualify.

Users must explicitly enable AI descriptions for their cameras to access the feature, and Google emphasizes that it is not designed for real-time security monitoring. The company frames the update as a convenience tool for routine automation rather than a replacement for dedicated security systems.

Looking ahead

While the feature is in early access, Google's integration of Gemini into home automation signals a growing trend toward AI-powered ambient computing. Future updates may expand compatibility to more camera models and regions, though no timeline has been provided. For now, the focus remains on refining accuracy and user experience within the constraints of the current preview phase.