Google’s Gemini Pet Memory Stumbles in Cat Recognition

Google is bringing Gemini into the smart home with Pet Memory, a feature designed to help Nest cameras distinguish individual household animals rather than merely detect that a pet is present. The system is intended to remember specific pets and enable tailored alerts or automated responses. Its performance matters because persistent visual recognition is central to Google’s effort to move Gemini beyond conversational services and into everyday home monitoring, where users expect identities to remain accurate over time.

A recent hands-on test found that Pet Memory misidentified cats and failed to retain identities it had previously learned, undermining its promise of personalized responses. The report did not disclose a launch date, accuracy rate or a numerical breakdown of the errors, limiting any broader assessment of performance. Still, the failures point to gaps in Gemini’s image recognition and long-term memory consistency, capabilities Google will need to strengthen before Pet Memory can be considered dependable for routine Nest camera use.