Why the freight still moves by hand
Will AI take over warehouse jobs? Not in one step, and not in the same way for every task. The core of this job is physical: moving freight, stock and materials to and from loading docks, delivery vehicles, containers and production areas. Someone has to break down a floor-loaded trailer, lift mixed cases, stack them so they don’t shift, and sort cargo before it goes on the next truck. Software can plan that work. It cannot carry it.
The hard part for machines is variation. One pallet holds shrink-wrapped cartons; the next holds sacks, loose tires and a broken crate. Loads settle in transit. Labels face the wall. Mobile robots now carry shelves and totes across flat, mapped floors, and they do it well, but reaching into a packed trailer and judging what to grab first is a different problem. That gap is why the robotics tier for this job stops at machines that move things rather than machines that handle them. Our guide to humanoid robots and physical work covers where that line sits.
Scale matters too. The Bureau of Labor Statistics counted about 2,950,280 of these jobs in the United States, with median pay of $40,240 a year (BLS, 2025). BLS projects employment to change by 1.8% between 2025 and 2035. That is slow growth, not decline. The realistic pressure here is fewer new openings in heavily automated sites, not whole crews disappearing.
What software runs, what it assists, what stays manual
Paperwork is the part machines already handle. Recording the number of units handled, matching shipments to work orders and updating counts in a warehouse system are steps a scanner and inventory software complete without a person writing anything down. Tasks in that group account for 0% of task time.
Assistance shows up in sorting and labeling. Systems decide which cargo gets staged first, print the tags that go on containers and route a cart to the right aisle, while a person does the sorting and attaching. That shared group covers 19% of task time. The judgment calls stay with the worker: what is damaged, what is mislabeled, what will not stack.
Everything else is still a person’s hands. Loading and unloading by hand, securing and bracing loads, assembling crates and containers, and keeping aisles and docks clear sit in the group that needs a human, which is 81% of task time. Our coverage score, which measures the share of task time AI can handle today, reads 7 out of 100. You can read how coverage is measured before reading much into it.
What the evidence actually shows
There is no direct test of AI against people doing this job. The quality-parity grade is D, and a D grade means not measured, so we publish no parity number for material movers. Benchmarks that compare models with professionals are written around screen work: drafting, coding, analysis. None of them unload a trailer.
What would settle it is operational, not academic: published throughput data from mixed-SKU trailer unloading, piece-pick error rates on unsorted freight, and injury and downtime figures from sites running mobile robots next to people. Vendor demonstrations are not that. Until independent numbers exist, treat claims that machines already match a loader on real freight as untested. Our full scoring method explains how grades change when better evidence arrives.
When the work could shift
Most likely after 2044 (8 in 10 of our scenarios). The replacement-year method explains what that window is built from.
Two things could pull it earlier. Grasping has improved fast, and a robot that can pick odd, soft and unlabeled items would hit the biggest manual block in the job. Large fulfillment operators also build new sites around machines from the start, so automation spreads quickest where ecommerce fulfillment volumes justify the capital.
Two things hold it back. Most buildings in use were not designed for robots: uneven floors, tight dock doors, no charging layout. And the cost case is weak in older sites with seasonal peaks, because a machine has to be paid for whether volume comes or not. People are hired and released with demand. You can follow how fast firms are adopting these systems in the business AI adoption tracker.
How to stay needed on the dock
Lean into the parts of the day that are not repeatable. Trailer unloading and hand loading of mixed freight is the clearest one. Securing and bracing loads so nothing moves is the second, because the cost of getting it wrong is damage and injury. Building and repairing crates and containers for awkward goods is the third; those items are exactly what automated lines reject.
Two skills raise your floor. First, certification on powered equipment: forklifts, pallet jacks, reach trucks. Second, being the person who clears exceptions in the warehouse system, because mismatched counts and bad labels land on whoever can read the screen and the pallet at the same time.
What to do: ask your supervisor which tasks the site plans to automate next, then get trained on the equipment that works alongside it.
Nearby work worth comparing: stockers and order fillers, hand packers and packagers, and industrial truck and tractor operators. For the wider picture, see the material moving workers family and the warehousing sector, or put two of these jobs side by side with the job comparison tool.