Google’s Pet Memory forgot who my cats are

| Source: The Verge AI

Tags: Google, Gemini, Nest, smart home, computer vision, pet recognition

Google's Gemini for Home Pet Memory feature, tested for two weeks on three cats, still cannot distinguish individual pets from one another — making personalized smart home automations like pet-specific feeding triggers unreliable in practice.

Details

Google's Pet Memory promises that Nest cameras integrated with Gemini for Home can learn to recognize individual household pets by name — going beyond generic animal detection to trigger pet-specific automations. In theory, this means a camera could identify which cat is at the food bowl and schedule feeding accordingly. After two weeks of testing with three cats (Boone, Smokey, and a new kitten named Osa), reviewer Jennifer Pattison Tuohy found the system consistently failed to distinguish between them. The cats received the same generic animal notifications rather than individualized ones, making the feature's core promise undeliverable in a multi-pet household. The failure is particularly significant because the ability to distinguish individuals is a prerequisite for the downstream automations Google advertises — feeding schedules, door automation, and tailored notifications. Without reliable identification, Pet Memory defaults to the same level of functionality as any standard motion-triggered camera. As a product review from a trusted outlet, this raises real questions about whether the feature is ready for its intended use case. Google has not publicly commented on the limitations observed.