EXPOSED
Operating continuous haulage, shuttle cars, LHDs and scoops in a low-seam underground environment is physical work that language models cannot touch — but it is the one blue-collar niche where purpose-built autonomy is already commercial. Autonomous and teleoperated loading/haulage systems (AutoMine, Cat MineStar, tele-remote scoops) are running in production mines precisely because removing humans from the face is a safety win, and the repetitive tram-load-dump cycle on fixed haul routes is the easiest case for machine autonomy. The occupation's protection is that most US underground operations are small, coal-heavy, capital-constrained, and shrinking for market reasons faster than they are automating.
Dipped in 2020, then grew past where it started.
Median pay $54,210 → $74,500 +9.9% 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
-22.3% 6,400 → 5,000 on the projections basis
Shrinking, but not obviously because of AI
The BLS projects -22.3% by 2034, but at 53/100 this work is only moderately exposed — not the profile of a job current AI can simply do. Occupations shrink for many reasons, and the score does not point at automation as this one's cause.
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.
~500 openings a year on average, including replacing people who leave.
MinerLoaderBuggy ManCar DumperJoy LoaderCar DropperCar PincherCart DriverCoke LoaderMuck HaulerYard LoaderBuggy DriverBuggy RunnerJoy OperatorMuck OperatorBuggy OperatorMachine LoaderMiner OperatorShuttle DriverLoader OperatorMucker OperatorShuttle OfficerProduction MinerRam Car Operator
Holding it up: embodiment . Weakest point: trust premium .
Tasks largely resist digitisation Tramming a scoop through a 42-inch coal seam, feeling out soft bottom, spotting loose roof between bolt rows and repositioning cable and ventilation curtain are tasks no software has replaced at scale, but the load-haul-dump cycle on a fixed haul route is the single most automated task in mining worldwide — AutoMine-class systems already run it in large hard-rock stopes, which is what pulls this off 17-18 and down to 14.
Hands-on in uncontrolled environments You work kneeling or seated in a low-profile machine in an unsupported-roof zone with respirable dust, methane, water on the bottom, changing ground conditions and no line of sight beyond the headlights — an uncontrolled environment in the literal MSHA sense, short of 20 only because the machine cab imposes some structure on the day.
Certification preferred, not legally required MSHA Part 48 training, annual refresher, task training on each machine and a state-issued miner's certificate in states like Kentucky and West Virginia are all mandatory, but none of it is a professional licence you personally lose in a liability action — the operator gets cited alongside the mine, and the operator holds the legal duty under 30 CFR 75.
Meaningful discretion You make real calls with no procedure covering them — whether that top will hold long enough to make one more trip, when to stop and pull the section boss over, whether the bottom will take the load — but the cut plan, roof-control plan and ventilation plan are written and approved above you, which caps the discretion below the ambiguous-high-stakes band.
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 (14/20 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 (4/20) is whether buyers specifically pay for a person. Judgment and accountability (9/20) 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 20 of this occupation's 53 points (38%).
Embodiment (19/20) 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.
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 67/100 — SAFE.
Task-mix shift as fixed-route tramming automates: the residual job becomes recovery of stuck/derailed machines, operating in irregular pillar-retreat and low-seam geometry, first cuts in newly developed entries, and manual haulage where roof conditions are deteriorating. This tier genuinely exists and resists autonomy because haul routes are not fixed. Rises only in surviving operations — headcount falls even as per-worker resistance rises.
MSHA rulemaking or an approval condition requiring a certified, MSHA-credentialed operator physically present in the section (not a surface teleoperation pod) to authorize and supervise autonomous haulage in gassy/permissible Class I coal environments — analogous to the existing permissibility approval regime for face equipment and the certified-person requirements for roof control and gas checks. If autonomy approvals are written to require an on-section certified attendant per machine rather than one remote supervisor per fleet, headcount protection follows.
UMWA or state-level (WV, KY, PA) mine safety board provisions conditioning autonomous haulage on a named certified operator retaining stop authority and bearing statutory duty — West Virginia's separate state mine certification regime is a live channel, and contract language on remote/automated equipment staffing has appeared in mining CBAs abroad.
If MSHA or state rules formalize the operator's role as the person who calls a halt on ground/roof/ventilation conditions observed during haulage — making that an accountable, documented determination rather than an informal report to the foreman — the role owns a consequential ambiguous call.
The limit. Even with every lever, the sector-level ceiling is low: US underground coal employment is contracting on commodity economics regardless of automation, and no trust-premium route exists — buyers purchase tons, never a human operator. Liability shields that require an on-section certified human are the only mechanism with real headcount consequence, and vendors are actively lobbying the opposite way, since removing humans from the face is the safety argument for autonomy.
| Beckley, WV | 200 | $74,130 +0% |
| Wheeling, WV-OH | 140 | $81,100 +9% |
| Tuscaloosa, AL | 120 | $57,730 -23% |
| Baton Rouge, LA | 40 | $47,910 -36% |
| Charlotte-Concord-Gastonia, NC-SC | 30 | $80,480 +8% |
| Wheeling, WV-OH | 140 | $81,100 +9% |
| Charlotte-Concord-Gastonia, NC-SC | 30 | $80,480 +8% |
| Beckley, WV | 200 | $74,130 +0% |
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 53. 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.