Amazon Tests AI-Powered Automated Delivery Stations

Amazon Explores Fully Automated Last-Mile Facilities

Amazon.com Inc. is developing fully automated delivery stations that would use artificial intelligence and robotics to handle the final stage of its logistics network, according to internal documents obtained by Business Insider. The initiative, code-named Project Tetromino, targets the sorting and staging operations that currently prepare packages for loading onto delivery vans.

Delivery stations represent the last stop in Amazon's supply chain before packages reach customers. Unlike fulfillment centers where items are picked and packed, these facilities receive already-packaged goods and organize them by delivery route. Automating this step would mark a significant shift in how the company manages its last-mile operations.

Why It Matters

Amazon's logistics infrastructure is a competitive moat that enables its two-day and same-day delivery promises. Automating delivery stations could dramatically reduce labor costs while increasing throughput during peak periods. The move also signals how AI and robotics are moving beyond warehouse picking tasks into more complex sequencing and loading operations that require spatial reasoning and real-time optimization.

Investment Timeline and Performance Targets

Internal planning documents reviewed by Business Insider show Amazon intended to invest $103 million in an initial Tetromino pilot program in 2028. The company would then build five additional sites in 2029 at approximately $85 million each, followed by 10 more locations by 2030. Total investment would exceed $530 million by 2029 if the plans proceed as outlined.

The automated facilities are projected to process packages at roughly 2.5 times the rate of Amazon's current delivery station designs. This performance gain would come from AI-driven sequencing that organizes packages in optimal delivery order before they reach human drivers.

Technology Partners and the Tetris Connection

Business Insider reports that Amazon may partner with Boxbot, an AI and supply chain robotics startup, for key technology components. Boxbot's system transfers packages from conveyor belts to storage trays and uses AI to sequence them for efficient loading. The startup claims its automated approach can accelerate the vehicle-loading phase by up to 10 times compared to manual methods.

The project's internal code name references Tetris, the classic puzzle game, reflecting the spatial optimization challenge of fitting packages into delivery vehicles in the correct sequence.

Broader Automation Push

The Tetromino project aligns with Amazon's aggressive automation expansion. During a July earnings call, the company announced plans to more than double its robotic arm fleet in 2026. Amazon's logistics competitors are pursuing similar strategies: FedEx has deployed AI-powered Dexterity robots for autonomous trailer loading, while UPS and DHL are adopting robots for unloading operations.

Company Response

An Amazon spokesperson told Business Insider that Project Tetromino remains an "early-stage concept" and disputed the accuracy of the leaked financial projections and timelines. The representative confirmed, however, that Amazon continues testing new technologies aimed at improving employee safety and customer experience.

Amazon stock gained 1.4% on Monday and extended gains in after-hours trading. The company's shares have risen 13% year-to-date.

These details were first reported by Business Insider.