Why a bridge or lock still needs someone on shift
A bridge tender works where road traffic and boat traffic meet. The job is to watch the channel, answer a horn or a radio call, hold back cars and pedestrians, raise or swing the span, then put everything back. Lock tenders do the same balancing act with water: open the gates, fill or empty the chamber, and move vessels through in order. Will AI replace lock tenders doing that work? On the evidence we score, most of the task time still sits with a person on site.
The reason is not that the machinery is clever. It is that the decisions are local and physical. A cyclist ducks under a lowered gate. A barge drifts off line in wind. A hydraulic line leaks, or the span sticks in cold weather and has to be worked by hand. Someone has to see it, stop the movement, and be accountable for what happens next. Public agencies and port authorities put a named operator behind those calls.
That does not mean the work is frozen. Remote and semi-automated control already exists on some structures, and where it is installed one operator can watch several bridges or locks from a control room. That is task erosion rather than a job disappearing: fewer booths, fewer first jobs, more monitoring per person. Federal figures show about 3,040 people employed in this occupation in the United States, with median pay near $57,700 and a projected decline of 2.3% between 2025 and 2035 (BLS). Small occupations move on budgets and infrastructure programs as much as on technology.
What software takes, what it assists, and what stays with the operator
The paperwork side is where automation lands first. Traffic and vessel counts, shift logs, opening records and maintenance notes can be captured and filed by software instead of written up by hand. Our share of task time AI can take on its own is 0%.
Assistance is the bigger story. Camera feeds, vessel tracking, gate and span sensors, and scheduling tools tell the operator what is coming and flag a vehicle still sitting on the deck. The share of task time where AI supports a person rather than replacing them is 19%. The operator still presses the button and still watches the water.
The rest belongs to people: clearing and checking the span or the lock chamber in person, hand-operating machinery when power or hydraulics fail, and judging when a movement is safe with pedestrians, cars and boats all waiting. That share is 81%. Taken across every task, our coverage score for this job is 14 out of 100, and how coverage is measured explains what that figure counts.
What has actually been tested
Not much, and we say so. Our evidence grade here is D. A grade of D means there is no direct, published test of an AI system against qualified bridge or lock tenders on their own tasks, so we publish no parity number for this job. Exposure estimates that rank occupations by task mix are not the same thing as a head-to-head trial.
What would settle it is specific: a documented comparison at structures of similar type and traffic, running staffed operation against remote or automated control, and reporting opening delays, vessel and vehicle incidents, equipment faults, and the staffing actually needed per structure. Until something like that is published, the honest answer is that the hands-on share is high and the proof is thin. Our quality parity method sets out how grades move once evidence arrives, and the full scoring method covers the rest.
When the picture could change
Most likely after 2044 (8 in 10 of our scenarios). The replacement-year method explains what that window describes and how it is built.
Two things could pull it earlier. The first is capital spending: when an agency rebuilds a movable bridge or a lock, automated gates, interlocks and remote control are easiest to design in from the start. The second is consolidation of control rooms, which spreads one operator across several structures without changing the machinery at all.
Two things hold it back. Over half the task time is physical, and the robotics tier this work would need is mobile robots, which is a harder and costlier class than a fixed arm in a factory. And the cost comparison shown above cuts both ways: software is cheap, but retrofitting old spans, gates and safety interlocks is not, and public agencies carry the liability if a sensor misreads a clear deck.
What to do: if your structure is being rebuilt or tied into a remote control center, ask now which tasks move to the console and who keeps the on-site certification.
How to stay needed in this work
Lean into the parts of the shift that need a person present. Keep the on-site safety judgment sharp, including how you clear pedestrians and vehicles before a movement. Stay the one who can run the span or the lock by hand when automation or power drops out. And keep the inspection habit: noticing a worn cable, a slow gate or a leaking seal before it stops traffic.
Two skills travel well. One is control-room monitoring, including reading camera feeds, vessel tracking and alarm systems for more than one structure at a time. The other is basic maintenance and troubleshooting of hydraulics, motors and electrical interlocks, which keeps you useful whether the booth stays or the work moves to a console.
Nearby work is worth a look. Traffic Technicians sits in the same part of the transportation family and leans on similar signal and traffic judgment. On the water side, Motorboat Operators and Dredge Operators both reward hands-on vessel and equipment skill. You can also browse the other transportation workers family, see how employers in transportation and warehousing are scored, or put two titles side by side with the job comparison tool.
Our Still needs a human score for this job is 80 out of 100 (higher is safer). To see where that sits against everything else, open the full job rankings or the list of jobs least exposed to AI.