Why lifting work stays with people
A hoist or winch job is judgment under load. The operator runs levers, pedals and brakes while a load hangs over ground, water or people. A rigger signals by hand or radio. The wind shifts, the line stretches, the load starts to swing, and the next move has to be right the first time. Nothing about that sequence is scripted.
The checks around the lift are just as physical. Operators inspect cables, ropes, drums and brakes for wear before a shift, and attach or release loads, pulleys and blocks at the hook. Software can flag a reading from a sensor. It cannot run a gloved hand along a wire rope or decide that today’s ground is too soft for the planned pick.
Scale matters too. The Bureau of Labor Statistics counts roughly 2,600 hoist and winch operators in the US, with median pay of $56,450 and projected employment change of about 0.2% from 2025 to 2035 (BLS, 2025). Automation money tends to follow big headcounts and repeat motions first. This job has neither.
What AI does, helps with, and leaves to the operator
No task on the list above sits in the group where AI runs the work on its own. The share AI handles without a person is 0%. That is what our first question, Can AI do it?, measures for this occupation.
The assist column is also empty for now, so no named task here is scored as shared work between software and the operator: 0%. In practice, digital tools still show up around the edges of the shift, in load monitoring, maintenance logging and scheduling, but those sit outside the tasks this score is built from.
Everything else is work people do. Moving loads with hoists, winches and drum-and-cable systems stays with the operator, and so does reading signals from riggers and spotters while the load is in the air. That share is 100%. The pattern explains the headline number on this page: 87 out of 100 (higher is safer).
What the evidence shows
There is no direct test of AI against a working hoist or winch operator. Our evidence grade for this job is D, which is the mark we use when the comparison has not been measured, so no parity number is published here. That is a gap in the research, not a finding in either direction.
What would settle it is specific. A timed field trial of an automated or remote winch system against an experienced operator, on real loads, recording placement accuracy, near-misses, emergency stops and recovery after a fault. Published operator-hours per lift from port, mine or offshore programs that already run automated hoisting would help too. Until something like that exists, the honest position is that the machine side has not been shown to match a person on this work. You can read how we grade and hold back parity scores in our scoring method.
When this could change
Most likely after 2044 (8 in 10 of our scenarios). What that window is measuring is explained on the replacement-year method page.
Two things could pull the date earlier. The first is equipment: remote-controlled and semi-automated hoisting already runs in some mines, ports and offshore setups, and the sensors, cameras and controls behind it keep getting cheaper. The second is standardization. Where a site lifts the same load along the same path every day, the pick becomes programmable, and the operator’s role shifts toward supervision.
Two things hold it back. Lifting is a regulated, liability-heavy activity, and certification and site rules still assume a qualified person is in charge of the load. And the equipment itself is long-lived. Plenty of winches and hoists in service have no data interface at all, so automating them means a retrofit, paid for against a small crew of operators.
Good to know: remote operation is not the same as automation; moving the operator into a cab across the yard still keeps the decisions with a person.
How to stay needed on the hook
Lean into the parts of the job that carry responsibility. Take the lead on pre-lift inspection of wire rope, drums and brakes, so you are the person who can say why a line was taken out of service. Own signal work with riggers and spotters, including radio discipline on noisy sites. And get good at non-routine picks, the awkward loads, confined spaces and bad ground that no fixed program handles well.
Two skills travel well from here. One is rigging and load calculation, including weight, center of gravity and sling angles. The other is working with monitoring and remote-control systems, including what to do when a readout and your own eyes disagree.
If you are weighing nearby work, compare the task mix with Crane and Tower Operators, Industrial Truck and Tractor Operators and Conveyor Operators and Tenders. You can also put any two jobs side by side on our job comparison tool, see the wider group on the material moving workers family page, or look at hiring and exposure across construction work. For a broader view of jobs built around physical judgment, our list of jobs that mostly need a person is the place to start.