Google Home Gemini introduced the ability to trigger automations based on visual insights from cameras, such as identifying people or objects, starting August 16, 2026 [1]. This feature is available in English to US users enrolled in the Google Home Public Preview program and supports Nest cameras plus select third-party models with Gemini Built-In [1].
Users can describe the events they want to trigger automations for using natural language. Google explained, "You just need to use simple, natural language prompts to describe the exact event you want to start an automation, and select which cameras should look for it" [2]. The company said this lets "your smart home automatically react to almost anything happening around your home" [1].
The feature requires enabling AI descriptions on compatible cameras and a subscription to the Google Home Premium Advanced plan, priced at $20 per month or $200 per year [1].
Google's Gemini AI first launched for smart home devices in early access in October 2025, replacing Google Assistant on those devices [1, 2]. The company has since improved Gemini’s ability to understand multiple voice commands simultaneously and accept more casual language [2]. The updated Google Home app also simplifies adding Familiar Faces and submitting feedback [2].
At Google I/O in May 2026, the Gemini built-in program expanded to more compatible cameras and speakers across 19 countries [2]. A new Google Home smart speaker powered by Gemini was announced in October 2025 and expected to ship in spring 2026 but has not yet become available for order as of August 2026 [2].
Additionally, Apple Music subscribers can once again stream the service on Google Home devices [2].
The camera-based automation feature marks the latest expansion of Gemini’s integration into smart homes. Users enrolled in the public preview program can test the automation capabilities starting now [1].
