EXPOSED
Driving a train on a fixed guideway with signal-based speed control is the most automatable form of vehicle operation, and fully driverless metros (Copenhagen, Vancouver, Dubai, Paris Line 1) already run at scale — the technology risk here was settled years ago. What protects US operators is not the task but the context: strong transit unions, decades-old signaling that must be replaced before automation is possible, and the political cost of removing a uniformed human from a car full of passengers. The surviving work is incident response, evacuations, door and platform judgment in crowded conditions, and reporting track or equipment defects.
Roughly flat across the period, with year-to-year wobble.
Median pay $67,880 → $86,380 +1.8% in real terms
This line is counted by the Bureau of Labor Statistics — the one figure on this page that isn't a judgement of ours. Headcount moves on demand, offshoring, demographics and the business cycle, and automation is one term among several, often not the loudest.
So a falling line is not evidence that AI did it, and a rising one is not evidence that it won't. Both happen in this register: some occupations resist automation and shrink anyway, others are highly automatable and keep growing. The marked year is 2020.
BLS projection, 2024–2034
+3.4% 9,600 → 9,900 on the projections basis
Exposed, but growing
AI can already do a lot of these tasks, and the BLS still expects +3.4% more of these jobs by 2034. Demand for the output is growing faster than the work is being automated away — the mechanism BLS gives for software developers, and the combination people most often misread as an error.
Different clocks. The score is what current AI could do to this work today. The projection is how many of these jobs will exist in 2034. Everything between the two — how fast employers actually adopt, whether demand grows in the meantime — is why they can point opposite ways without either being wrong.
~900 openings a year on average, including replacing people who leave.
TrammerTram DriverRail OperatorRide OperatorTram OperatorMotor OperatorTrain OperatorOperator TrainerSubway ConductorTransit OperatorTrolley OperatorVehicle OperatorElectric MotormanElevated MotormanMonorail OperatorStreetcar MotormanStreetcar OperatorTrack Car OperatorLight Rail OperatorSubway Train DriverMotor Coach OperatorTrackmobile OperatorTrolley Car OperatorRail Transit Operator
Holding it up: embodiment . Weakest point: trust premium .
Mixed — a routine tier and a judgment tier Holding the throttle to a signaled speed profile on a fixed guideway is a solved problem — CBTC does it on the Victoria and Canarsie lines already — so the 8 rests entirely on the non-driving residue: berthing accuracy at curved platforms, reading a crowded platform edge before releasing doors, and getting a stalled train and 800 passengers off in a tunnel with third rail live.
Some physical or field component You are sealed in a cab in a controlled tunnel most of the shift, which caps this well below a track worker, but the 11 reflects the parts that put you outside it: walking the consist to cut out a sticking brake, climbing down to the roadbed for an evacuation or a person under the train, and manually cranking or bypassing doors when the pneumatics fail.
Certification preferred, not legally required You need a federally recognized certification and route qualification (49 CFR Part 243, and FRA Part 240-style certification where the line is regulated) plus rules-book recertification, and you can be decertified and drug-tested after a signal overrun — but you sign no design or dispatch decision and the authority carries the liability, which is why this sits at 7 rather than in the licensed-professional range.
Meaningful discretion The rulebook covers most of your shift — flagging, restricted speed, absolute block, radio protocol — and dropping to 8 rather than lower is warranted because the calls that aren't scripted are big: whether to hold in the tunnel or run to the next platform, when to key by a red signal, whether an unresponsive passenger stops the train.
The verdict above describes this occupation as a whole. Almost nobody does the typical version of a job — tick what's actually in your week and see how your own mix sits.
Your task mix speaks to task resistance (8/20 here) — how much of the day's work current AI already does. That is the dimension the boxes above are about.
It cannot move the other three. Liability shield (7/20) is whether the law requires a licensed human to sign. Trust premium (6/20) is whether buyers specifically pay for a person. Judgment and accountability (8/20) is whether the role exists to own consequential calls. Those are facts about the occupation's standing, not about which tasks are in your week — a paralegal who does only trial exhibits still holds no licence. Together they are 21 of this occupation's 40 points (53%).
Embodiment (11/20) is also a property of the work rather than the worker, but we don't tag individual tasks as physical or not, so the picker can't tell you anything about it. That's a limit of this tool, not a claim.
Did we get the list right? Tell us what's missing — the tasks are written from the outside, and you're reading this from the inside.
Locomotive Engineers EXPOSED
Commercial Pilots SAFE
The moves above are yours to make. This is the other half: what would have to change in the world for the occupation itself to score higher. None of it is in any one person's gift, but it is where the floor actually comes from. Scores here are not a one-way ratchet. Only two of the five dimensions — task resistance and embodiment — track what machines can do. The other three track law, what buyers will pay for, and who is answerable, and those move in both directions, often in response to the same pressure AI creates. If every lever below landed, this occupation would score around 58/100, still EXPOSED.
Task-mix shift under CBTC: as speed and stopping control move to the signal system, the retained role becomes train attendant/incident commander — owning evacuation calls in tunnels, third-rail power-removal requests, and bypassing failed door interlocks. If agencies formalize this as a certified 'train operator/incident response' classification with sign-off authority (as some GoA2 metros already do), the ambiguity ownership rises.
Same two-tier split: the routine tier (throttle, braking, station stops) is what CBTC eats; the residual tier is smoke/fire response, ADA boarding assistance, de-escalation with disruptive passengers, and defect reporting on track and rolling stock, which no deployed system handles. Task_resistance measured against the residual job rises even as headcount falls.
A negotiated or legislated crew-size floor for automated operation — e.g. TWU Local 100 contract language or a state law of the kind New York considered requiring a credentialed attendant aboard every train, or an FTA State Safety Oversight Program Standard conditioning GoA4 approval on a certified onboard employee responsible for evacuation. This converts the operator into a legally mandated presence rather than a driver.
Rider-facing safety concern (assaults, homelessness response) driving agencies to keep uniformed staff aboard for perceived security, as in transit-ambassador programs at BART and LA Metro. Note this rewards a presence, not the operating skill, and is a budget line agencies can cut.
The limit. Every lever here protects a body in the car, not the job of driving; headcount can fall sharply even where the classification survives, and the capital cost of resignaling is the real clock — once CBTC is installed, the operating task is gone regardless of what the contract says.
| New York-Newark-Jersey City, NY-NJ | 3,950 | $90,140 +4% |
| Seattle-Tacoma-Bellevue, WA | 320 | $77,020 -11% |
| Atlanta-Sandy Springs-Roswell, GA | 270 | $59,570 -31% |
| San Diego-Chula Vista-Carlsbad, CA | 200 | $60,970 -29% |
| Baltimore-Columbia-Towson, MD | 150 | $84,350 -2% |
| Minneapolis-St. Paul-Bloomington, MN-WI | 130 | $69,680 -19% |
| Virginia Beach-Chesapeake-Norfolk, VA-NC | 70 | $61,980 -28% |
| St. Louis, MO-IL | 40 | $63,030 -27% |
| New York-Newark-Jersey City, NY-NJ | 3,950 | $90,140 +4% |
| Baltimore-Columbia-Towson, MD | 150 | $84,350 -2% |
| Seattle-Tacoma-Bellevue, WA | 320 | $77,020 -11% |
We have no reported case of a named organisation automating this occupation. Not one deployment, not one announcement.
That is worth saying out loud next to a score of 40. The verdict above is about what the work exposes — what current AI could do to these tasks. It is not a claim that anyone has done it. For this occupation those two things have come apart completely: the capability argument is on this page, and the evidence column is empty.
Has AI actually changed your work? One tap, anonymous, and the running tally is public. Nothing else is asked of you.
Rather than check back: get the digest and we'll tell you what changed — or watch a single occupation from its own page.