The Arpalus platform utilizes computer vision to monitor store shelves with an accuracy rate exceeding 95%. This technology enables Instacart’s 600,000 shoppers to capture granular inventory data during order fulfillment, providing a live feed of product availability. Beyond manual shopper inputs, the system will integrate directly with Caper smart carts, which collect automated data as customers navigate store aisles.

This move deepens Instacart’s "Physical AI" strategy, which seeks to bridge the gap between brick-and-mortar operations and e-commerce. David McIntosh, the company's chief connected stores officer, noted that the goal is to create a unified ecosystem where in-store reality mirrors digital storefronts instantly. The company currently leverages machine-learning models trained on 1.6 billion lifetime orders and live data feeds from nearly 100,000 North American retail locations. Financial terms of the acquisition remain undisclosed.