Feeding multiple cats sounds simple until you have one food-aggressive tabby that hoovers up a diabetic housemate’s prescribed meal before you’ve even left for work. Petlibro thinks it has finally cracked that problem. The pet-tech company has unveiled an AI-powered smart feeder designed specifically for multi-cat homes, using on-device machine learning to identify individual animals and gate their access to food — no microchip implant required. It is the kind of AI infrastructure thinking applied at the smallest possible scale, and it could genuinely change daily life for millions of pet owners.
According to a TechCrunch report published September 19, 2026, the device uses a built-in camera and a trained vision model to distinguish between cats in the same household, triggering the lid to open or stay closed depending on which animal approaches. The system learns each cat’s facial and body features during an initial setup period, then operates autonomously without a cloud connection for core recognition tasks.
How the Recognition System Actually Works
The feeder’s AI model runs locally on the device, meaning it does not depend on a continuous Wi-Fi signal to decide whether to dispense food. That matters in practice: a dropped connection at 6 a.m. will not leave your cats staring at a locked bowl. Petlibro says the system can differentiate between cats even when they look physically similar — a common failure point for older RFID-based feeders that rely on collar tags or implanted chips that not all owners want to use.
Each feeder can be enrolled with multiple cat profiles, and portion sizes are set individually per animal through a companion app. That means a vet-prescribed low-calorie plan for one cat and an unrestricted feeding schedule for another can coexist on the same device without constant human supervision. The company has not yet disclosed the exact size of the training dataset used to build the vision model, but it claims high accuracy even under low-light kitchen conditions — a meaningful spec given that many cats eat at dawn or late at night.
A Real Market Problem Gets a Hardware Answer
Multi-pet feeding has been an underserved category in smart home hardware. Most connected feeders on the market treat every approaching animal identically, making them functionally useless for households where cats have different dietary needs — which, according to veterinary nutrition guidance, is extremely common as pets age or develop conditions like kidney disease, hyperthyroidism, or obesity. The demand for a genuinely individualized solution has existed for years; the hardware and on-device AI have only recently become cheap enough to make it viable at a consumer price point.
Petlibro is pricing the new feeder competitively within the premium smart feeder segment. The company has built a reputation for reliable, app-connected pet devices, and this launch extends that brand into territory where software differentiation — not just build quality — becomes the core value proposition. It is a trajectory worth watching as AI capabilities continue to migrate from data centers into everyday household objects, often solving mundane but genuinely frustrating problems that previous generations of connected devices were too blunt to address. For anyone managing a household where the cats are on competing meal plans, that is not a small thing.
