Example: I analyzed Symbotic this week (using AI). They design, build, and deploy AI-driven robotic warehouse systems for Walmart and others. Their robots de-palletize inbound freight into individual cases (SKUs), chaotically stow those cases wherever space is open, and then rebuild store-ready pallets in the exact sequence needed for each truck—even down to packing pallets by the specific store aisle where they’ll be shelved. That boosts space utilization by creating very densely-packed warehouses, minimizes robot-traffic bottlenecks, and speeds retrieval.
What AI does here simply couldn’t be programmed in advance—no one could predict which spaces would be open or what retrieval conditions would apply. That’s why Symbotic uses AI: it dynamically maps every available stowage space, anticipates future demand, and tracks bins in real time to direct robots accordingly. What looks like chaos is actually continuous optimization: i.e., “thinking.”
I’m a computer scientist and yes, I’ve tested several of the most popular models. I stand behind my comment.